Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Sustainable Consumption and the Intention–Behavior Gap

Abstract. Sustainable consumption and the intention–behavior gap concern the relation between prospective environmental or ethical purchasing intentions and choices made under actual market constraints. This fundamental research monograph synthesizes the history of environmental behavior science, the theory of planned behavior, moral norms, consumer identity, habitual purchasing, constrained preference implementation, and contemporary evidence on pricing, availability, labeling, product performance, social influence, and market design. It distinguishes attitude–intention, intention–action, and action–environmental-impact discrepancies; evaluates survey and transaction measurement, field experiments, and meta-analyses; and compares findings from food, ethical certification, durable goods, household services, and digital shopping. Particular attention is given to the interpretation of fair-trade field experiments, social-norm replication, implementation-intention research, meta-analytic evidence of small and heterogeneous environmental intervention effects, and persistence after interventions end. The synthesis concludes that stated commitment and observed purchases are linked through mechanisms whose expression depends on real opportunity, prices, routines, information, and institutions. It proposes research designs for longitudinal calibration, causal identification, distributional evaluation, and artificial decision-support systems. The work is a critical narrative synthesis of published science; it does not claim original participant data, a preregistered systematic review, or a new quantitative meta-analysis.

Keywords: sustainable consumption; intention–behavior gap; green purchase intentions; environmental consumer behavior; ethical consumption; planned behavior; behavioral intention; actual purchasing; sustainability labeling; consumer choice; habit formation; green product availability; behavioral interventions; social norms; field experiments; sustainable purchase persistence; Artificial Era

The Scientific Object of Sustainable Consumption

Sustainable consumption concerns the acquisition, use, maintenance, sharing, and disposal of goods and services in ways intended to reduce environmental burdens and, in some research traditions, social harms. The object studied here is narrower and experimentally tractable: the relation between consumers' stated or measured intentions to choose a more sustainable option and their subsequent purchasing behavior when options, prices, information, and opportunities are specified. A consumer can endorse environmental protection while purchasing a conventional product, purchasing nothing, delaying replacement, or adopting a lower-impact practice outside retail markets. These outcomes require separate treatment. Environmental consequences ultimately depend on production systems, lifetime use, substitution, and disposal as well as what the consumer chooses. Accordingly, selecting a product labeled green is a behavioral indicator under particular conditions, not a self-validating measure of ecological benefit. The present dossier follows the economic and psychological sequence from preferences to actual choices, with particular attention to observations of market behavior Stern (2000) White et al. (2019).

The intention–behavior gap is a difference between a prospective disposition to act and an observed action in a defined opportunity window. It becomes measurable when intention specifies an action, target, context, and temporal horizon corresponding to the recorded outcome. Agreement with a general statement such as "I care about the environment" is an attitude, whereas intending to buy a particular certified product on the next shopping trip is an action-oriented measure. A reported gap may represent motivational inconsistency, practical barriers, changing preferences, inaccurate prediction, deficient recall, socially desirable responding, or the researcher's mismatched measures. The construct therefore comprises a family of explanatory problems rather than an individual failure of character. In a retail encounter, availability, inventory, time, money, and product performance can determine whether an initially preferred product becomes the chosen one Ajzen (1991) Sheeran & Webb (2016).

The central questions are whether sustainable intentions predict objectively observed purchases, which mechanisms condition the relationship, how causal interventions can improve actual outcomes, and how durable effects remain under repeated transactions. We examine primary experiments, field studies, longitudinal observations, conceptual models, and quantitative syntheses. Each design supplies a different inference: a cross-sectional survey establishes covariation; an incentivized choice experiment identifies responses under a controlled choice set; a randomized retail intervention can estimate the effect of changing information or prices on transactions within the study setting. This is a critical narrative synthesis, not a preregistered systematic review or newly calculated meta-analysis. Evidence quality is judged by behavioral validity, experimental identification, sample selection, baseline constraints, temporal follow-up, replication, and whether the consumption outcome is linked to a credible environmental impact Lange & Dewitte (2019) Osbaldiston & Schott (2012).

Sustainable Products, Practices, and Impact Measures

A sustainable consumption decision can involve food sourcing, a durable appliance, transportation services, household energy, clothing, packaging, or repair instead of replacement. Those decisions differ in frequency, cost, social visibility, and the consumer's control over outcomes. A reusable bag or an environmentally labeled beverage is a relatively frequent and low-stakes choice; home insulation is infrequent, capital intensive, dependent on housing tenure, and mediated by finance and contractors. Treating these outcomes as interchangeable obscures mechanisms. Studies of household conservation can illuminate habit and social influence, but their findings do not automatically estimate sustainable purchasing effects. Research on energy feedback, for example, studies usage reductions under repeated information, while research on fair-trade labels studies the demand response to a visible ethical attribute Allcott (2011) Hainmueller et al. (2015).

The environmental criterion itself demands definition. Some papers infer sustainability from certification, organic status, fair-trade claims, local production, plant-based categories, energy ratings, or perceived environmental friendliness. Each operationalization captures different properties. Fair-trade certification foregrounds supply-chain labor and trading standards, while lifecycle climate emissions may depend on production, transport, use, and substitution. An energy-efficient appliance can have lower operating demand per use yet produce little net savings if use increases or a still-functioning product is prematurely replaced. Researchers should specify whether they measure the choice of an attribute, expenditures on a labeled option, total resource consumption, product lifetime, or environmental impact. The choice of denominator also matters: purchases per eligible opportunity, share of units, share of expenditure, and annual household impact answer different questions Steg & Vlek (2009) White et al. (2019).

Sustainable consumption is further shaped by supply-side institutions. Product assortment, retailer placement, labeling standards, warranty systems, repair services, infrastructure, public policy, and relative prices affect the available action space. Even a highly motivated consumer cannot choose an absent product, and even a nominally affordable option may carry search, travel, learning, and verification costs. Behavioral explanations that omit availability may incorrectly assign a structural constraint to intention weakness. A useful causal representation distinguishes consumer capacities and motivations from opportunity structures, then studies how their interaction determines feasible choice. This approach makes product design and market access legitimate units of psychological investigation while retaining observed purchasing as the defining outcome of this dossier Ertz et al. (2016) Reisch & Zhao (2017).

What Counts as an Intention–Behavior Gap?

The apparent size of a gap depends on how researchers pair intention and action. A question asked months before purchase and phrased broadly about green products is weakly compatible with a receipt documenting one specific purchase. Stronger designs collect intentions shortly before a defined transaction, observe whether the relevant product is stocked, record the person's budget and alternatives, and match the outcome to the stated behavior. The familiar distinction among attitude, intention, and conduct is therefore substantive, not grammatical. Vermeir and Verbeke's study of sustainable food examined an attitude–behavioral-intention relation; it provides insights into intention formation but should not be cited as direct proof that respondents failed to purchase sustainable foods, because actual transactions were not its central dependent variable Vermeir & Verbeke (2006) Ajzen (1991).

Three separable gaps are useful in research design. The attitude–intention gap asks whether evaluations become plans; the intention–action gap asks whether specified plans lead to behavior; the action–impact gap asks whether behavior produces the expected environmental or social consequences. A survey may identify the first gap, a loyalty-card panel the second, and lifecycle-informed consumption accounting the third. The three can coincide, but one cannot be inferred from another. A person may intend to consume sustainably and execute that intention while the selected product offers little verified impact; alternatively, a person may reduce resource demand without identifying with green consumption. The meaning of a successful intervention must therefore be specified before interpreting statistical associations Stern (2000) Lange & Dewitte (2019).

Intention can also change between measurement and choice. New information about price, quality, household needs, or financial security may rationally revise preferences. In such cases, a discontinuity between the earlier intention and later purchase is compatible with responsive decision-making. Other gaps arise through automatic selection, inattention, inaccurate beliefs, choice overload, default options, or anticipated social judgment. Distinguishing revised intention from failed implementation requires repeated measures across the decision sequence and, where possible, process evidence such as search paths, exposure to availability, and transaction timestamps. The general intention–behavior literature emphasizes that even strong intentions frequently remain unrealized and that the degree of correspondence depends on control, stability, and action specification Sheeran & Webb (2016) Gollwitzer (1999).

Origins of Environmental Behavior Research

The contemporary gap literature emerged from several traditions that should be interpreted together. Early environmental education research related knowledge, attitudes, personality, and situational conditions to responsible behavior. Hines, Hungerford, and Tomera synthesized empirical studies in the 1980s and identified relationships between psychosocial variables and environmental conduct while exposing the limitations of simple knowledge-to-action accounts. The turn toward social-psychological prediction brought intention and perceived control into sharper focus. Ajzen's theory of planned behavior formalized behavioral intention as a proximal predictor influenced by attitude, subjective norm, and perceived behavioral control, while preserving a role for actual control. The resulting framework allows researchers to ask why people who endorse environmental values might act differently when money, opportunity, or performance requirements intervene Hines et al. (1987) Ajzen (1991).

Around the early 2000s, integrative environmental research increasingly treated behavior as an interaction of motives and contexts. Stern distinguished environmentally significant behavior by its impact and by its intent orientation, warning that intent is not identical to environmental consequence. Kollmuss and Agyeman synthesized barriers to pro-environmental action across internal and external domains and popularized the problem of a mismatch between concern and behavior. Later meta-analyses by Bamberg and Möser and by Klöckner combined intention, norms, habit, and contextual variables in formal comparisons. These developments are important because they replaced a single-sequence information model with a layered account in which different pathways can become decisive across settings Stern (2000) Kollmuss & Agyeman (2002) Bamberg & Möser (2007) Klöckner (2013).

Consumer-specific work then moved from general environmental conduct toward the ecology of retail decisions. Vermeir and Verbeke analyzed sustainable food intentions in relation to attitudes and perceived availability; Young and colleagues used qualitative accounts of green consumers' purchasing experiences; Carrington, Neville, and Whitwell offered a framework for why ethical purchase intentions may be lost during implementation. Field experiments on fair-trade labeling, product framing, and retailer environments added behavioral outcomes to the predominantly self-report literature. The historical lesson is methodological: every increase in behavioral realism changes what can be concluded. A study measuring stated willingness to pay, a controlled laboratory selection, and supermarket sales may all investigate sustainable choice, yet they estimate distinct outcomes and operate under different constraints Vermeir & Verbeke (2006) Young et al. (2010) Carrington et al. (2010) Hainmueller et al. (2015).

The Theory of Planned Behavior and Its Conditions

The theory of planned behavior proposes that intention is shaped by evaluation of an action, perceived social expectations, and perceived capacity to perform it. Intention should predict performance most closely when the relevant action is under volitional control and the measures share a common target, context, and time frame. Perceived behavioral control can contribute to intention and, when it represents actual constraints, directly to behavior. Sustainable product purchasing provides a critical test: consumers can believe a certified item has social value and feel pressure to choose it while confronting a price premium or missing shelf stock. Such circumstances separate motivational commitment from genuine control Ajzen (1991) Bamberg & Möser (2007).

The model is often applied through surveys that measure attitudes, normative beliefs, perceived control, and behavioral intention. These predictors can explain variation in stated readiness to buy an environmentally preferable product, but a high model fit for intentions is not a validation of actual purchases. Reports from a single questionnaire share wording, response style, and social-desirability pressures. More persuasive studies pair baseline measures with later receipts, point-of-sale observations, or repeated transaction records, and test incremental prediction beyond prior purchase habits, price, and store availability. Comparisons should use predictive calibration as well as association: a statistically reliable correlation may still produce poor identification of who purchases the sustainable option Vermeir & Verbeke (2006) Sheeran & Webb (2016).

An additional limitation is the assumption that the opportunity structure remains sufficiently stable after intention measurement. Market offers and household obligations change. A person's expectations can be inaccurate, or the behavior can require collective coordination with others. A rigorous consumer study should therefore distinguish confidence in one's ability, external feasibility, and realized opportunities instead of placing all three into a single perceived-control item. This distinction supports experimental questions: does reducing the price premium improve behavior among equally motivated participants, and does perceived availability mediate willingness to seek information? Such designs can identify mechanisms that a self-report path model alone cannot establish Ertz et al. (2016) Reisch & Zhao (2017).

Values, Personal Norms, and Environmental Identity

Theories of environmental action emphasize moral obligation and identification alongside instrumental cost-benefit assessment. The norm-activation tradition concerns personal responsibility and the perceived consequences of conduct; value-based approaches distinguish concern for ecological systems, other people, and one's own interests. In sustainable consumption, these orientations may contribute to the salience of environmental attributes before a purchase is made. However, a biospheric value orientation does not guarantee that green attributes dominate price, quality, familiarity, and household needs at every decision. An adequate account asks whether values change attention, attribute weighting, perceived obligation, or the threshold for accepting tradeoffs, and whether that pathway survives observation of actual choices Stern (2000) Steg & Vlek (2009).

Environmental identity concerns the extent to which environmental consideration becomes part of a person's self-understanding and enacted routines. It may stabilize behavior across changing situations, or it may produce symbolic choices that are especially attractive when publicly recognizable. Experiments by Griskevicius, Tybur, and Van den Bergh found that status motives can favor visibly green products under particular social conditions, demonstrating that ecological claims may interact with reputation and audience. Their findings illuminate a mechanism rather than a universal rule of consumer motivation. Researchers should record public versus private settings, product visibility, local norms, and whether the purchase has a verifiable material benefit Griskevicius et al. (2010) White et al. (2019).

Ethical consumption also includes tensions between moral conviction and practical self-description. Consumers can neutralize a disappointing purchase by invoking cost, skepticism about labels, or the apparent insignificance of one action. Qualitative studies are valuable for identifying these explanations as lived reasoning, but they cannot independently determine which factor caused a transaction. Integration with purchase panels and experimental changes in costs or claims can test whether reported justifications correspond to actual constraints. The empirical challenge is to distinguish sincerely held environmental identity, anticipated social approval, and retrospective rationalization without treating any one as a complete theory of the gap Johnstone & Tan (2015) Carrington et al. (2010).

The Architecture of a Consumer Purchase Episode

Sustainable purchasing is a sequence of opportunities and decisions rather than a single psychological moment. Consumers first encounter a need or desire, consider whether acquisition is necessary, identify a feasible channel, search or recall available options, evaluate attributes, and act under time and budget constraints. Later use, repair, and replacement determine whether the purchase displaces a less sustainable alternative. This sequence suggests distinct failure points: an environmental preference may never enter the consideration set; an apparently attractive option may fail a minimum quality standard; or a prepared intention may be defeated by stockouts at checkout. Research that observes only the final selected product loses information about which step produced the discrepancy. Studies of real purchasing therefore benefit from linked data on attention, search, assortment, stated choice criteria, prices, and transaction records Young et al. (2010) Carrington et al. (2010).

A consideration-set account explains why attitudes measured outside the market can overstate the environmental attribute's practical role. A respondent can answer a survey by considering the ethical ideal in isolation, whereas a real shopping trip forces simultaneous comparison of price, durability, taste, convenience, and reputation. Attributes need not be combined through a single stable weighted average. Consumers may use noncompensatory rules such as refusing any product above a budget threshold or below a performance minimum. Environmental superiority then matters only within the subset of options that meet basic requirements. Comparing unrestricted hypothetical judgments with actual choices under imposed price and quality criteria can distinguish genuine motivational change from changes in the feasible set Luchs et al. (2010) Ertz et al. (2016).

Repeated purchasing introduces feedback. A successful sustainable choice can improve confidence, lower search costs, and create familiarity; a disappointing product can lead the consumer to reject a category for reasons unrelated to environmental motivation. Consumer experience is therefore part of the explanatory system rather than a postscript. Panel designs should record repeat purchase, product satisfaction, substitution, and switching, while separating first-time trial from long-term adoption. Interventions aimed at a single choice can produce an initial response that disappears after a price promotion or label disappears. The appropriate time scale depends on product replacement cycles and must be long enough to capture the proposed mechanism White et al. (2019) Agarwal et al. (2025).

Price Premia, Budgets, and Opportunity Costs

Price is an observable constraint with both economic and psychological components. A sustainable product may require a higher immediate payment while promising lower operating costs or greater durability. Consumers compare these benefits under limited cash, competing demands, uncertainty, and attention. A reported willingness to pay is influenced by the hypothetical framing and may exceed the amount paid when actual money is at stake. A credible test distinguishes stated willingness, incentivized willingness, posted retail prices, and realized expenditures. The same nominal premium can have different effects across income groups because it consumes different shares of discretionary resources. A study reporting low sustainable purchasing without measuring the feasible price differential cannot establish that intention failure was the cause Reisch & Zhao (2017) Hainmueller et al. (2015).

Budget constraints are dynamic. Food costs repeat weekly, whereas the efficiency gain of an appliance accrues over years. Consumers may face liquidity barriers even when lifetime savings are positive, and uncertain future benefits are especially difficult to compare with an immediate price. Researchers should identify who pays, who receives the benefit, how long the product will be held, and whether financing or split incentives intervene. Housing renters, for example, may have fewer opportunities to invest in durable improvements than owners. These conditions are partly institutional, yet their psychological consequences operate through perceived control, anticipated regret, risk, and the accessibility of alternatives. Sustainable-choice policy should be evaluated by observing whether changes to feasible prices and financing produce actual shifts in behavior Steg & Vlek (2009) Ertz et al. (2016).

The fair-trade coffee field experiment by Hainmueller, Hiscox, and Sequeira is instructive because it manipulated information in functioning retail settings and observed purchases. Ethical labeling affected demand, with responses sensitive to the product and price conditions, demonstrating that expressed support can translate into market behavior when a recognizable attribute is available. The experiment does not establish a general effect for all sustainable products, and fair-trade certification should be distinguished from climate impact. Its methodological strength is that it tests transactions rather than relying exclusively on moral agreement. Replication should cross income distributions, price-premium magnitudes, store formats, and competing product characteristics Hainmueller et al. (2015).

Availability, Search Costs, and Convenience

Availability determines whether intentions can be implemented. A highly motivated consumer may lack nearby stores carrying verified alternatives, may encounter an empty shelf, or may face a digital catalog in which environmental attributes are difficult to locate. Convenience includes travel, delivery, waiting, information verification, and cognitive effort. These costs are often invisible in surveys asking whether a respondent would choose a sustainable option if it were offered. Researchers should observe inventory and the actual choice set at the decision moment, not infer accessibility from respondent belief alone. Retailer-level assortment changes and randomized product placement can help establish whether easier access changes purchases among consumers with different baseline environmental intentions Vermeir & Verbeke (2006) Reisch & Zhao (2017).

Search becomes particularly difficult when sustainability claims are multidimensional. Labels can convey organic certification, recycled content, fair labor, energy efficiency, or carbon footprint, each with a different standard of evidence. A shopper may be uncertain whether two competing labels are comparable, and this uncertainty can lead to default reliance on familiar brands. Additional information can help, but more competing claims may also increase verification burden. The mechanism is an interaction between attribute comprehension and the time available for choice. Experiments should separately manipulate informational clarity, display prominence, and the credibility of the certifying institution, then observe what changes at checkout White et al. (2019) Young et al. (2010).

Convenience is also embodied in household routines. Shopping lists, delivery subscriptions, meal plans, and automatic replenishment lower the cost of repeating an established choice. An environmentally preferable substitute may therefore require intentional interruption of a stable workflow. This observation supports a distinction between interventions that make sustainable options easier to choose and those that demand continuous attention to moral goals. Changing the default assortment or reordering a catalog can shift behavior without changing underlying environmental attitudes. Whether the result persists depends on the stability of the new environment and on customers' satisfaction after use Klöckner (2013) Mertens et al. (2022).

Habits, Automaticity, and the Persistence of Consumption

Habit refers to learned tendencies for cues to elicit responses after repeated enactment in relatively stable contexts. Shopping routines can become habitual even when people retain strong values and explicit intentions. Brand selection, route through a store, preferred package size, and recurring digital orders reduce the number of attributes reconsidered on each occasion. A sustainable intention measured in a survey may remain accessible as an abstract goal without being retrieved at the decisive point. This mechanism predicts that the effect of intention on purchase varies with past-behavior frequency and situational stability. A repeated purchase is not by itself proof of psychological automaticity; researchers should use validated habit indicators, context manipulations, and process observations Klöckner (2013) White et al. (2019).

Change can occur when familiar cues are disrupted. Moving to a new residence, changing retailers, beginning a different commute, or altering a household budget can reopen previously automatic decisions. Researchers can exploit such changes in longitudinal designs, but these naturally occurring transitions also alter prices, available products, family circumstances, and social networks. A causal test should separate cue disruption from the new opportunity structure. Planning interventions can also link a specific cue to a selected action, for example identifying the shelf or shopping-list reminder that will prompt examination of a lower-impact alternative. Such techniques are better suited to behaviors within the person's control than to decisions blocked by unaffordable prices Gollwitzer (1999) Gollwitzer & Sheeran (2006).

Sustainable consumption may require replacement of a sequence rather than a single response. Buying less, repairing a product, or shifting to a reusable system involves procurement, storage, maintenance, and coordination over time. Success at initial acquisition can coexist with later abandonment if cleaning, refilling, or service access becomes burdensome. One-shot experiments therefore risk identifying curiosity or novelty rather than durable practice. Repeated purchase panels and utilization logs can reveal whether a purportedly successful green intervention changes the cumulative behavior that matters. The research focus should be persistence after initial enthusiasm and after subsidies, reminders, or public attention fade White et al. (2019) Agarwal et al. (2025).

Product Performance and the Sustainability Liability

Consumers rarely evaluate an ethical attribute in isolation. Products must also satisfy expectations about effectiveness, safety, reliability, taste, convenience, and appearance. Luchs, Naylor, Irwin, and Raghunathan documented a sustainability liability: under some conditions an ethical attribute can produce inferences that a product is gentler and potentially less strong or effective. This result illustrates attribute interaction, especially when strength is a valued functional dimension. The theory should not be generalized to all categories; the relevant research question is which sustainability claims change perceived performance, under which product goals, and whether those inferences alter actual purchase Luchs et al. (2010).

The implication extends beyond advertising. If a sustainable product is expected to perform worse, a consumer may rationally choose the conventional alternative while retaining environmental commitments. Comparative performance tests, real-use trials, product guarantees, and informative demonstrations can help resolve this tradeoff, but experimental studies must distinguish changed beliefs from changed purchases. Such a design can randomly vary performance information while holding price and environmental claims constant, then record both perceived utility and willingness to buy. Some behavioral effects may appear only after actual experience, making postpurchase satisfaction and repeat transactions essential to causal interpretation Luchs et al. (2010) Young et al. (2010).

Trust is another performance-like attribute. Consumers must often rely on claims they cannot personally verify, including material sourcing, emissions accounting, or supply-chain labor standards. Skepticism may arise from contradictory claims, unfamiliar certification, or memories of disappointing purchases. The psychologically relevant construct is the credibility of a specific claim in a particular decision, rather than an undifferentiated trait called green skepticism. Research can manipulate transparent verification or independent certification and observe whether improvements in comprehension and trust lead to actual substitution. The environmental impact of substitution should then be evaluated through separate product-level evidence White et al. (2019) Reisch & Zhao (2017).

Social Visibility, Reputation, and Moral Signaling

Purchasing can communicate identity to observers. An environmentally preferable product may signal responsibility, competence, group membership, or status, especially when its sustainable attribute is recognizable to others. Griskevicius, Tybur, and Van den Bergh used experiments to show that status motives can increase preference for green products under conditions of public visibility, including circumstances in which conspicuous conservation is socially valued. A public-versus-private setting is therefore not merely a detail of data collection; it changes the incentives attached to a product. These findings help explain why a consumer might choose a conspicuously green product in a visible setting but pursue a different tradeoff in private Griskevicius et al. (2010).

Normative messages can also alter how people interpret their own intentions. A descriptive norm tells consumers what similar others do, while an injunctive norm conveys what is approved. Their effects need not coincide, and a norm that is poorly matched to the relevant community may have weak credibility. In hotels, the original experiments by Goldstein, Cialdini, and Griskevicius reported favorable responses to descriptive conservation messages, whereas subsequent field work by Bohner and Schlüter challenged the reliability of particular comparative advantages. These studies concerned towel reuse rather than product purchase, but they reveal why social-norm mechanisms require context-specific replication and measured behavior Goldstein et al. (2008) Bohner & Schlüter (2014).

Visibility can introduce a measurement problem. People may make socially desirable claims when asked about ecological intentions, particularly in an interview or survey with moral cues. That response bias enlarges the observed attitude-to-purchase discrepancy even when true private motivation is weaker than reported. Anonymous prospective measures, indirect behavioral indicators, and confidential transaction linkage can test this possibility. The substantive account should distinguish reputation-driven purchases from survey impression management and from enduring environmental identity. Each predicts different responses when observers, anonymity, and choice costs are varied Carrington et al. (2010) Griskevicius et al. (2010).

Research Designs and the Hierarchy of Inference

The central inferential challenge is to measure environmental intentions and subsequent consumption using designs that preserve their temporal order. Cross-sectional surveys are inexpensive and can compare values, perceived availability, and intentions across groups, but they cannot reliably separate causal influences or observe later action. Prospective surveys improve temporal ordering yet remain vulnerable to selective participation, attrition, and recall bias. Experience sampling and shopping diaries provide repeated within-person information, though reporting can itself increase attention to sustainable choices. Administrative receipt or loyalty-card linkage is closer to actual purchasing, provided researchers address missing transactions at other retailers and the limits of household-level identification. The design should begin with a defined behavioral estimand rather than a convenient questionnaire item Lange & Dewitte (2019) Sheeran & Webb (2016).

Choice experiments manipulate attributes such as certification, price, and product performance under controlled conditions. Their value depends on incentive structure and ecological realism. Hypothetical selections allow factorial investigation of competing attributes but can exaggerate morally approved responses. Incentivized experiments impose real costs and can reveal tradeoffs more accurately, yet participants may encounter an artificial assortment or a one-time decision. Retail field experiments and natural experiments capture behavior within functioning markets but may face imperfect randomization, contamination, seasonal change, and attrition. A triangulated evidence strategy asks whether conclusions converge across these designs rather than declaring one universally superior Hainmueller et al. (2015) Osbaldiston & Schott (2012).

A strong causal design specifies assignment, exposure, compliance, behavioral outcome, follow-up interval, and an analysis plan before treatment. For a price-label intervention, the relevant estimate could be an intention-to-treat effect on the share of eligible purchases, with heterogeneity by baseline intention. To identify mediation through belief or attention, additional assumptions or experimental manipulations are required. A single statistically significant difference in posttest attitude is insufficient to establish behavior change. Context-specific interventions should be compared with plausible counterfactuals, including no intervention, a neutral informational message, or an economically meaningful alternative Nisa et al. (2019) Mertens et al. (2022).

Measurement, Psychometrics, and Transaction Validity

Intention measures should correspond to target, action, setting, and time. Asking whether consumers plan to purchase any sustainable products over the next year does not identify the same action as buying certified coffee during a particular grocery trip. Measures can specify the product category, expected opportunity, willingness to incur a premium, and confidence that the chosen option will be available. Reliability matters, but consistent wording does not cure mismatch between intention and observed outcome. In particular, general environmental concern and self-reported green identity should be treated as potential predictors, not interchangeable operational definitions of intention. The classic planned-behavior framework is most informative when behavior compatibility is designed into the measurement protocol Ajzen (1991) Vermeir & Verbeke (2006).

Behavioral measures involve their own errors. A receipt may show purchase but not who used the product or whether it replaced another purchase; a loyalty card may omit cash transactions, competing retailers, returns, and purchases by other household members. Consumer diaries capture unobserved contexts but invite memory errors and impression management. In observational panels, prior purchase share can help account for stable differences, though adjusting for prior behavior may obscure pathways through which past interventions changed habits. Researchers should report sample recruitment, store coverage, missingness, transaction definitions, and whether an indicator is validated against an independent record Lange & Dewitte (2019) Young et al. (2010).

Outcome magnitudes also require care. A percent increase in the share of green products is not equal to a comparable decrease in total environmental impact. The denominator could be total purchases, a category's purchases, dollars spent, or purchasing occasions. A low baseline can make a large relative percentage correspond to a small absolute change. For household-resource studies, energy consumption reductions should be converted to environmental outcomes only with assumptions about generation and usage; for products, substitution and rebound may matter. A rigorous report supplies absolute changes and uncertainty where available and keeps effect-size interpretation within the original outcome and population Allcott (2011) Nisa et al. (2019).

Sustainable Food, Organic Labels, and Ethical Coffee

Food purchasing is a productive setting for gap research because the choices recur frequently and involve familiar sensory, economic, and moral attributes. Sustainable-food intentions can arise from animal welfare, production methods, labor conditions, environmental concerns, or perceived health, each with different mechanisms. Vermeir and Verbeke's study of sustainable food linked attitudes and behavioral intentions to perceived availability and social factors. Its contribution is to intention formation, while the actual intention-to-purchase relation requires direct transaction evidence. Repeated grocery panels can test whether stronger stated intentions predict changes in category shares after accounting for household budget, product assortment, brand loyalty, and taste preferences Vermeir & Verbeke (2006).

Fair-trade labeling provides a particularly useful test of ethical claims in retail settings. Hainmueller, Hiscox, and Sequeira studied consumer demand with multistore field experiments that manipulated fair-trade information and observed sales. Their design establishes that an ethical attribute can influence behavior under the study's pricing and retail conditions. The results should be interpreted as evidence about demand for fair-trade-certified products, not as direct estimates of reduced carbon emissions or as universal evidence that every ethical label changes purchases. Follow-up questions concern whether labeling induces sustained switching, whether shoppers understand the certification, and how effects depend on premium levels and product familiarity Hainmueller et al. (2015).

Food-category research also illustrates the difference between product substitution and consumption reduction. Selecting a sustainability-labeled product can preserve overall purchase quantity, whereas wasting less food or replacing high-impact meals could change total resource demand more substantially. Researchers should define whether the target is choice among offered labels, avoidance of a conventional attribute, reduction of total material throughput, or a verified change in environmental impact. Their interventions and behavioral indicators should be designed accordingly. Consumer psychology explains why choices are made, while lifecycle and supply-chain analysis are needed to determine whether the resulting consumption pattern meets the substantive environmental objective Steg & Vlek (2009) White et al. (2019).

Clothing, Electronics, Durability, and Replacement

Durable goods transform the intention–behavior problem because purchasing less may be more sustainable than buying a newly marketed green alternative. Clothing choices involve style, identity, quality, affordability, and anticipated social visibility. Electronics involve performance, software support, repairability, warranties, and rapidly changing product standards. A survey that measures willingness to buy sustainable products may miss the intention to extend the life of an existing item. Studies should therefore distinguish new sustainable acquisition, avoidance of acquisition, repair, resale, and product retention. These behaviors have different opportunity structures and do not necessarily share one psychological intention Stern (2000) White et al. (2019).

Durability also changes the time horizon over which consumer preferences can be assessed. The purchase of a product advertised as repairable is an immediate action; the later decision to repair it is contingent on defects, repair price, service access, and continued utility. A person may execute an environmental intention at purchase yet never face the future repair opportunity. Conversely, consumers may prolong an existing product's use for economic reasons without reporting environmental intentions. A credible empirical design can follow ownership duration, repair attempts, replacement triggers, resale, and product use, rather than treating sustainable brand preference as a complete measure of impact Steg & Vlek (2009) Ertz et al. (2016).

Functional tradeoffs are particularly important for categories in which performance dominates the choice. The sustainability liability identified by Luchs and colleagues suggests that ethical attributes sometimes activate unfavorable expectations about product strength. For durable goods, long-run reliability and confidence in manufacturers may outweigh immediate signaling. Randomized information trials can investigate whether independently verified performance or repairability information changes purchase and retention, while field evidence can examine whether warranty or repair-market reforms alter actual replacement. The essential principle is to match psychological explanation to the consumer's full ownership trajectory Luchs et al. (2010) Reisch & Zhao (2017).

Household Energy, Transportation, and Adjacent Behavior

Household energy and mobility are relevant comparators because they involve repeated environmentally consequential decisions, although many outcomes are usage rather than purchase. Allcott's field research on home-energy reports demonstrated how social comparisons can change household electricity use in large-scale settings. The outcome was consumption of energy, not the purchasing of an environmentally labeled product. Its value for this dossier lies in the experimental demonstration that feedback and comparison can shift real behavior when opportunities for adjustment exist. Similar designs could be applied to sustainable tariffs, appliance choice, or transport subscriptions, but their effects require independent testing in those purchasing settings Allcott (2011).

Transportation transitions illustrate structural dependence. An intention to choose low-emissions travel may be constrained by route coverage, travel time, safety, disability access, household obligations, and vehicle availability. The psychological meaning of perceived control changes across these settings. Choices between two transit products, for example, differ from a choice between transit and car travel in a city lacking reliable services. Researchers should combine intention measures with origin-destination feasibility, costs, observed trips, and the quality of available alternatives. An intervention may improve attitudes without changing behavior because it does not alter the feasible route set Stern (2000) Steg & Vlek (2009).

Adjacent domains also help identify what can and cannot be generalized. Hotel towel reuse is a pro-environmental action under a local norm manipulation, not a retail transaction. Home-energy feedback changes a utility-use pattern, whereas fair-trade labels target product choice. Evidence from one domain offers hypotheses about norms, habit, salience, and attention but cannot substitute for category-specific estimates of actual sustainable purchases. Reviewers should code outcome type and behavioral opportunity alongside intervention type, particularly when pooling disparate forms of environmentally relevant action Goldstein et al. (2008) Bohner & Schlüter (2014) Hainmueller et al. (2015).

Socioeconomic, Cultural, and Life-Course Variation

Sustainable purchase opportunities are distributed unevenly. Household income, local retail density, transport options, work schedules, digital access, and housing arrangements shape what products and services are realistically available. Price premia can impose substantially different burdens across households even when average willingness to pay appears favorable. Survey comparisons across income groups must therefore distinguish preferences from purchasing power and opportunities. The causal question is not whether one group cares more about the environment in the abstract, but which changes in affordability and access would alter actual choice for each group. Representative sampling and objective measures of local options are central to this inquiry Ertz et al. (2016) Reisch & Zhao (2017).

Culture enters through meanings attached to waste, thrift, status, responsibility, family consumption, and trust in institutions. An environmentally labeled product may communicate social esteem in one market and skepticism about elitism in another. Cross-cultural comparison requires measurement equivalence: questions about personal norms and purchase intentions must be understood similarly before mean differences are interpreted. Researchers should also document local certification systems, retail channels, and environmental policy, because the same survey wording can describe very different purchase opportunities. Multilevel designs can separate person-level beliefs from market- and region-level institutions, although causal conclusions require more than observed country correlations Stern (2000) Griskevicius et al. (2010).

Life-course circumstances influence both habit and control. Young adults may be forming consumption routines while facing limited income; parents coordinate purchases with other household members; older adults may have established brand habits and different mobility or technology needs. These are hypotheses about mechanisms, not universal age stereotypes. Strong studies compare people at similar opportunity levels or follow them through changes in employment, housing, household composition, and product access. Such designs help distinguish stable preference, cohort norms, and situational change. Equity-sensitive intervention design tests effects in populations actually facing the most restrictive market conditions Young et al. (2010) Klöckner (2013).

Information, Certifications, and Product Labels

Information interventions assume that some consumers would choose differently if relevant attributes were easier to identify or verify. Product labels can lower search costs, clarify the meaning of sustainability standards, and change which criteria are considered at checkout. Their effects depend on comprehension, trust, competition among claims, the visibility of price, and whether the consumer believes the category matters environmentally. A label can increase stated willingness to pay without increasing actual sales when price or quality constraints dominate. The fair-trade retail experiments conducted by Hainmueller, Hiscox, and Sequeira demonstrate the value of comparing labeled and comparison conditions using real transactions; the estimate applies to the manipulated claim and product context Hainmueller et al. (2015) White et al. (2019).

The content and framing of an appeal can change its interpretation. Hardisty, Johnson, and Weber experimentally examined different wording for a carbon-related payment and found that framing interacted with political affiliation in their study setting. This suggests that what consumers understand a policy or product surcharge to mean may influence willingness to pay beyond its numerical value. It does not establish a universal partisan effect, nor should findings about carbon offsets be treated as direct evidence for all green purchases. Scientific reporting should specify the wording, policy context, sample characteristics, and exact dependent variable to support valid translation Hardisty et al. (2010).

More information is not necessarily better information. A shelf filled with competing marks and technical language may increase decision effort, while a concise, credible claim can reveal a usable attribute. Researchers can separately randomize label visibility, independent verification, precision of environmental impact data, and claims about product performance. Behavioral outcomes should include substitution to the labeled product, total category purchases, and persistence after repeated exposure. When factual accuracy of a claim is contested, a psychological experiment cannot resolve its environmental truth; that requires a separate audit of the product or certification system. Evidence-based labeling joins behavioral responsiveness with verifiable substantive content Reisch & Zhao (2017) Luchs et al. (2010).

Choice Architecture, Defaults, and Retail Design

Choice architecture changes how options are presented without necessarily changing their underlying economic attributes. Placement, assortment order, standard selections, comparison units, and defaults can affect attention and selection. In digital retail, these elements include sorting, filtering, recommender-system ranking, and recurring-order settings. An intervention that makes a lower-impact option prominent may change behavior by improving retrieval rather than persuasion. A default can also increase adoption when active choice is costly. Evaluations should measure whether consumers notice the intervention, whether they can readily opt out, and whether any increase in sustainable choices represents genuine substitution instead of an addition to total spending Reisch & Zhao (2017) Mertens et al. (2022).

Nudge evidence is heterogeneous. The large meta-analysis by Mertens and colleagues reviewed choice-architecture interventions across behavioral domains, demonstrating that effects vary by architecture and decision context. Its pooled estimate is not a sustainable-consumption-specific result. The environmental experimental review by Osbaldiston and Schott likewise considered a mixture of behavior types and intervention components. This matters because effects on recycling or household energy conservation cannot simply be transferred to purchases of durable products. A dossier-specific conclusion requires coding objective outcomes, purchase context, intervention duration, and comparator conditions rather than assigning one general effectiveness score to all nudges Mertens et al. (2022) Osbaldiston & Schott (2012).

Ethical evaluation is intertwined with effectiveness. Choice architecture can support a previously stated consumer goal, but it can also steer people toward higher-priced products on the basis of ambiguous claims. Transparent disclosure, simple opting out, independent environmental verification, and attention to heterogeneous financial burdens are therefore scientific and normative evaluation criteria. A randomized change in default may be effective in altering observed selection while worsening welfare for a subgroup with different needs. Researchers should record product satisfaction, complaint or return rates, consumer comprehension, and distributional effects whenever feasible Reisch & Zhao (2017) White et al. (2019).

Social Norms, Feedback, and Replication

Social influence acts through beliefs about what others do, anticipated approval, reputation, and group identity. Environmental appeals often ask consumers to compare their choices with those of a local or relevant reference group. Goldstein, Cialdini, and Griskevicius tested descriptive norms in hotel towel reuse and found evidence that particular norm messages could influence behavior in that setting. The finding encouraged an influential program of norm-based interventions but addressed reuse rather than product purchase. For consumer goods, an analogous hypothesis requires field tests using true relevant peer behavior and direct purchase measurement Goldstein et al. (2008).

Replication qualifies how general an intervention appears. Bohner and Schlüter examined hotel towel reuse in field work and did not reproduce the original comparative advantage for some descriptive norm appeals. Differences among hotels, populations, messages, measurement practices, and statistical uncertainty can change outcomes. The broader principle is to distinguish existence of a context-bound effect from a claim that a mechanism works reliably in every marketplace. Preregistered replications with clear randomization and outcome definitions help establish which aspects are transportable Bohner & Schlüter (2014) Goldstein et al. (2008).

Large-scale household feedback provides complementary evidence. Allcott analyzed energy-conservation reports using randomized assignment across many households and found modest average reductions in electricity use. Such work offers an example of actual behavior tracked repeatedly and at scale. Yet energy-use feedback is not equivalent to purchasing green products, and spillover from energy use to retail choice is a new empirical question. Interventions should be evaluated on their intended outcome; a reviewer can learn about social comparison and persistence while preserving the distinction between behavioral domains Allcott (2011) Nisa et al. (2019).

Implementation Intentions and Goal Enactment

A motivation to purchase sustainably may remain too abstract to shape an action at the checkout point. Implementation intentions are conditional plans that specify when, where, and how a response should occur. Gollwitzer proposed that such plans can link anticipated cues to intended behavior, thereby reducing reliance on spontaneous deliberation. A consumer might identify a concrete shopping occasion and decide in advance to compare two certified products once they become available. The plan remains dependent on product availability, budget, and sufficient information, which separate the psychological initiation mechanism from economic feasibility Gollwitzer (1999).

Gollwitzer and Sheeran's meta-analysis demonstrated substantial implementation-intention effects across multiple goal domains. This result supports conditional planning as a general intervention candidate; it does not establish an equally large effect for sustainable shopping in real markets. Sustainable-consumption trials need to identify a specific purchase behavior, randomize action planning against an appropriate comparison, and obtain receipts or verified transactions. Outcome measurements should include plan recall, encounters with planned cues, execution when the opportunity actually arose, and durability beyond the experimental reminder Gollwitzer & Sheeran (2006) Sheeran & Webb (2016).

Planning can interact with environmental design. A reminder before entering a store may work when the intended option is present and easy to find, but fail when the product is unavailable or requires a prohibitive premium. In some settings, implementation intentions may target an alternative strategy, such as repairing rather than replacing, arranging access to a borrowing service, or reducing avoidable purchases. Testing these mechanisms requires comparing planning alone, improved opportunity alone, and their combination. A factorial experiment can estimate whether cue-based preparation and environmental support operate independently or synergistically without presuming the answer Gollwitzer (1999) Reisch & Zhao (2017).

Meta-Analytic Evidence, Effect Magnitudes, and Durability

The evidence synthesis must distinguish the gap itself from interventions designed to close it. Bamberg and Möser synthesized psychosocial predictors of pro-environmental behavior, emphasizing intention, perceived control, moral norms, and related influences. Klöckner integrated environmental-behavior models with attention to habit and behavioral control. These works characterize patterns of association across varied ecological behaviors; they do not calculate a universal rate at which sustainable purchase intentions become transactions. A consumer-specific effect estimate requires samples whose outcomes are actual purchases of clearly defined products under observed opportunities Bamberg & Möser (2007) Klöckner (2013).

Randomized experimental syntheses offer more direct evidence about changing behavior. Osbaldiston and Schott examined experimental pro-environmental interventions across multiple settings. Nisa, Bélanger, Schumpe, and Faller synthesized randomized controlled trials of household climate actions and reported small average behavioral effects and limited evidence of persistence after intervention. These conclusions challenge the expectation that inexpensive informational or psychological interventions will reliably transform environmentally consequential behavior by themselves. However, household climate-action studies include more than retail purchases, and their pooled results should be used as a boundary on optimism rather than an exact estimate for sustainable-consumption markets Osbaldiston & Schott (2012) Nisa et al. (2019).

The strongest interpretation of a behavioral intervention requires absolute impact, statistical uncertainty, baseline prevalence, follow-up, and environmental relevance. A statistically detectable increase in labeled-product purchases might be important in a large market, but consumer substitution, rebound, and product-level environmental performance determine eventual ecological benefit. Meta-analytic averages can conceal effect modification by price sensitivity, retailer, country, product category, and measurement duration. Publication bias and correlated treatment arms can distort a pooled estimate when not handled appropriately. Policymakers and researchers should therefore prefer transparent, domain-specific evidence tables over a single headline effect Nisa et al. (2019) Mertens et al. (2022).

Publication Bias, Replication, and Causal Quality

The sustainable-consumption literature combines laboratory experiments, convenience-sample surveys, field trials, and observational consumer panels. Each design creates different vulnerabilities. Survey common-method variance can inflate the association between environmental attitudes and intentions; socially desirable responding can overstate willingness to purchase; underpowered studies may exaggerate the magnitude of successful interventions. Field trials can suffer from treatment contamination, unobserved promotional changes, or differences among stores. A credible critical synthesis records the unit of randomization, sampling frame, outcome source, statistical adjustment, preregistration where available, and whether the reported effect is the main planned comparison Lange & Dewitte (2019) Osbaldiston & Schott (2012).

Replication should recreate the causal mechanism with realistic variation in context. Repeating a label experiment in a different product category tests generalizability, while repeating the same label and product in new stores tests sensitivity to market setting. Null findings can identify limits, as illustrated by the divergent hotel social-norm results. A well-specified null result is scientifically informative when confidence intervals rule out effects large enough to matter in practice; a nonsignificant test in a small sample often remains inconclusive. Both favorable and unfavorable studies should inform a judgment about practical intervention strength Bohner & Schlüter (2014) Goldstein et al. (2008).

Causal mediation requires special care. A treatment that changes purchases and also changes reported concern does not prove that concern mediated the transaction effect. Post-treatment measures can be jointly affected by treatment and participant expectations; changes in store stock can confound the inferred mechanism. Experimental manipulation of information, price, or availability, along with temporally ordered process measures, can help distinguish alternative explanations. The resulting evidence should answer specific questions such as whether perceived price causes an intention shortfall in a defined market rather than invoking a universal psychological barrier Reisch & Zhao (2017) Hainmueller et al. (2015).

An Integrative Model of Constrained Sustainable Consumer Action

The evidence supports a layered explanation of sustainable purchasing in which values and evaluations generate preferences, intentions specify prospective action, and transaction environments determine whether those intentions can be implemented. Perceived social norms, identity, and anticipated performance influence the formation and salience of intentions. Money, stock availability, product quality, time, and household coordination constrain feasible action. Habit and attention govern which attributes are considered at the choice point. Feedback from actual use then updates preferences and routines. This is a conceptual integration of established research programs, not an experimentally validated new structural model. Its value lies in identifying variables that should be measured separately and intervention pathways that can be tested against observed market outcomes Ajzen (1991) Stern (2000) Klöckner (2013) White et al. (2019).

The model makes competing explanations discriminable. If environmental intention predicts purchase only when the price premium is low, affordability moderates translation into action. If a label changes choice only when prominently displayed, attention or comprehension may limit implementation. If sustainable purchasing occurs mainly in public, social signaling may contribute. If purchase persists after reminders stop, habit formation or continued consumer satisfaction may be involved. Each explanation should generate prospective predictions that survive measurement of stock, price, prior purchases, and changing intentions. Factorial experiments can test interaction among motivational prompts, feasible options, and friction removal Griskevicius et al. (2010) Hainmueller et al. (2015) Gollwitzer & Sheeran (2006).

The framework also clarifies why there is no single universal gap percentage. Different studies observe different products, opportunities, time windows, and definitions of intention. A more coherent quantitative program would report calibration curves linking intention strength to observed purchases for a specified opportunity set, alongside subgroup estimates and uncertainty. It could then compare the added predictive value of price, inventory, prior behavior, and identity. This approach respects both the psychological aspiration of explanation and the economic reality of constrained choice, while keeping environmental impact assessment as a related but distinct inferential stage Sheeran & Webb (2016) Lange & Dewitte (2019).

Causal Inference Across Markets and Institutions

A consumer's choice takes place in a market that determines the cost and availability of alternatives. Retailers select products, public institutions regulate claims, producers design goods, and financial institutions shape payment options. These factors can affect both what consumers expect and what they ultimately buy. Observational differences between green and conventional buyers may therefore arise partly from neighborhood access, marketing exposure, income, or product availability. Multilevel analyses can describe these relationships, while randomized assortment, pricing, or labeling changes offer better leverage for estimating specific causal effects. Claims about a psychological gap should be evaluated against these alternative accounts Ertz et al. (2016) Reisch & Zhao (2017).

Natural experiments can complement randomized trials when a retailer changes its assortment, a jurisdiction introduces labeling, or a service becomes newly accessible. Difference-in-differences or related quasi-experimental comparisons require credible assumptions about trends, selection, concurrent shocks, and spillovers. A policy implemented in a high-income district may not yield transferable effects where alternatives remain expensive. Consumers and retailers may also adapt to the intervention, changing relative prices or product placement. Researchers should report implementation fidelity and market response along with behavioral outcomes rather than treating the intervention as isolated from the economic system Allcott (2011) Hainmueller et al. (2015).

The causal quantity of interest may be the effect of making one option available, reducing its premium, communicating a credible attribute, or combining interventions. These questions target different mechanisms. An information campaign is unlikely to correct a supply shortage; a subsidy may increase purchases without changing environmental identity; a default can change selections even when respondents' stated preferences remain stable. Successful design begins by identifying the binding constraint, then measuring whether its removal changes verified behavior and downstream consequences. Such an approach avoids mistaking a weakly predictive intention item for a sufficient account of market outcomes Steg & Vlek (2009) Nisa et al. (2019).

Equity, Consumer Welfare, and Ethical Evaluation

Environmental goals interact with consumer welfare. Policies that promote sustainable products can alter household expenditures, convenience, privacy, and access to essential goods. A premium that is trivial for one household can be prohibitive for another, so aggregate effectiveness can conceal unequal burdens. Studies should stratify effects by relevant economic circumstances and test whether changes in supply, subsidies, or service provision improve participation. An ethical approach does not assume that consumers who fail to purchase expensive alternatives lack commitment. It investigates the actual opportunity set and the distribution of intervention benefits and costs Ertz et al. (2016) Reisch & Zhao (2017).

Consumer autonomy also matters. Nudges and recommendation rankings influence attention and selection; their legitimacy depends partly on truthful claims, accessible alternatives, transparency, and the ability to decline. Labeling based on environmental claims of uncertain validity risks directing purchases without achieving environmental benefit. Research evaluations should therefore record comprehension, satisfaction, returns, and unintended substitution, not simply the treated group's click-through or purchase share. Where an intervention uses personal transaction histories, privacy and representativeness become additional design requirements. Evidence of behavioral influence is one component of policy judgment, alongside verified impact and consumer welfare Mertens et al. (2022) White et al. (2019).

Distributive analysis is especially important when households are asked to act on collective environmental problems through individual purchasing power. Some high-impact choices are constrained by housing tenure, public transport infrastructure, income, or product regulation. Interventions can reduce decision friction, but institutional change may alter the choice set more substantially. A scientifically responsible synthesis therefore locates individual behavior within its institutional setting and tests coordinated changes in price, availability, and information. This prevents the intention–behavior gap from becoming a rhetorical explanation for outcomes that are driven primarily by material constraints Stern (2000) Steg & Vlek (2009).

Digital Commerce, Ranking Systems, and the Artificial Era

Digital commerce makes sustainable consumption increasingly dependent on artificial information environments. Search rankings, recommendation systems, default subscriptions, and personalized offers determine which products enter the consideration set and which attributes become salient. A shopper may express an environmental intention while encountering a feed optimized for conversion, low immediate price, or previous purchases. These systems can reduce search costs if sustainability information is accurate and visible, or amplify friction if credible options are buried. The psychological mechanisms remain attention, perceived control, price comparison, habit, and trust; the technical architecture changes how these mechanisms are activated White et al. (2019) Reisch & Zhao (2017).

Artificial agents and consumer decision-support systems create a new empirical question: can an automated service translate a person's stated environmental goals into consistent choices while respecting budgets, product performance, and genuine preferences? A controlled evaluation could compare human-only shopping, recommendations that explicitly account for independently verified sustainability attributes, and automated filtering against a fixed consumer-approved constraint set. Dependent variables should include purchased products, spending, satisfaction, override behavior, false sustainability claims, and real environmental impact. Such a proposal belongs to the research agenda of the Artificial Era, not the established findings of earlier shopping psychology. It requires transparency about who selects attributes, whose preferences are represented, and how recommendations are verified.

The transition From Homo to Artificial broadens the study from individual decision processes to relations among human intentions, computational rankings, producers, and market institutions. Artificial systems can modify opportunity and information structure before a person reaches the final choice. This makes system-level exposure an important causal variable that should be recorded alongside attitude and intention. Studies of ranking changes and automated advice must distinguish improved goal implementation from manipulation or simple channel switching. The same methodological standard applies: clear intervention, defined counterfactual, observed purchasing, uncertainty, and independent verification of ecological claims Mertens et al. (2022) Lange & Dewitte (2019).

Unresolved Scientific Questions and Research Program

The first unresolved question concerns measurement. Research needs datasets in which prospective intentions, genuine options, actual prices, inventory, and longitudinal transactions are observed together. Such datasets would make it possible to estimate the share of apparent intention failure attributable to changing goals, missing opportunities, inaccurate beliefs, habits, and measurement error. A particularly useful protocol would collect an intention before a shopping trip, log the available products and their verified attributes, record the actual purchase, and reassess intention after new information becomes available. Preplanned subgroup analyses could assess how the resulting relation differs by income, previous purchasing, and product category Ajzen (1991) Lange & Dewitte (2019).

A second priority is intervention durability and impact. Many experiments evaluate an immediate response to a label, norm message, or altered default. Fewer establish persistence once the intervention is removed or whether observed substitution yields lifecycle environmental gains. Future trials should record repeated purchases, total consumption, returns, longer-run satisfaction, and environmental consequences independently of consumer self-description. Registered protocols and public analytic plans would also improve the credibility of null results and cross-site replication. Meta-analyses should distinguish real transactions from attitudes and intentions and avoid pooling behavior types whose mechanisms and denominators are incompatible Nisa et al. (2019) Mertens et al. (2022).

A third priority is institutional heterogeneity. Experiments that jointly change product access and psychological intervention can identify whether motivation is a limiting factor once economic constraints are relaxed. Cross-country studies should measure local markets rather than treating national identity as an explanation. Digital-commerce studies should record rank order, search exposure, personalized pricing, and recommendation constraints. These research designs would permit theories of environmental identity and behavioral control to compete against availability, transaction costs, and choice architecture as explanations of actual outcomes. This agenda also creates opportunities to evaluate the role of artificial decision-support systems through reproducible, human-centered market experiments Ertz et al. (2016) Reisch & Zhao (2017) White et al. (2019).

Conclusion

Sustainable consumption and the intention–behavior gap constitute a specific research problem at the intersection of consumer psychology, economic choice, and environmental behavior. The central finding of the literature is that favorable attitudes and sincere intentions are insufficient to predict an observed purchase without knowing the feasible product alternatives, costs, performance requirements, and decision context. The theory of planned behavior helps explain intention formation and behavioral control; norm, identity, habit, and information models explain additional pathways. Field experiments and transaction-based designs demonstrate that changes to labeling and choice conditions can influence actual behavior, while replication studies and meta-analyses show that effects are heterogeneous and often modest Ajzen (1991) Hainmueller et al. (2015) Nisa et al. (2019).

The scientific synthesis separates attitude–intention discrepancies, intention–action discrepancies, and action–environmental-impact discrepancies. These are distinct estimands requiring different data. Survey intentions are informative about motives, actual transactions about behavior, and independently verified product and lifecycle evidence about environmental consequences. A rigorous study aligns action and time-frame definitions, measures stock and price, records actual opportunities, and distinguishes association from causality. The strongest intervention evidence comes from realistic comparison conditions, repeated observation, transparent effect estimates, and independent replication Lange & Dewitte (2019) Osbaldiston & Schott (2012) Sheeran & Webb (2016).

The long-term research task is to explain how consumers execute environmental intentions within the markets they inhabit. Progress depends on investigating motives and constraints together: how people form goals, how products become visible and credible, how habits structure repeated buying, and how institutions determine affordable and practical options. Artificial commerce and decision-support systems add a new layer of opportunity shaping whose effects can be studied with the same experimental discipline. The resulting science concerns the actual organization of sustainable action, its measurable conditions, and the consequences of changes in that organization for both consumers and the environment Stern (2000) White et al. (2019).

References

  • Agarwal, P., Kumar, D., & Katiyar, R. (2025). Antecedents of continuous purchase behavior for sustainable products: An integrated conceptual framework and review. Journal of Consumer Behaviour, 24(4), 1685–1710. https://doi.org/10.1002/cb.2487
  • Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
  • Allcott, H. (2011). Social norms and energy conservation. Journal of Public Economics, 95(9–10), 1082–1095. https://doi.org/10.1016/j.jpubeco.2011.03.003
  • Bamberg, S., & Möser, G. (2007). Twenty years after Hines, Hungerford, and Tomera: A new meta-analysis of psycho-social determinants of pro-environmental behaviour. Journal of Environmental Psychology, 27(1), 14–25. https://doi.org/10.1016/j.jenvp.2006.12.002
  • Bohner, G., & Schlüter, L. E. (2014). A room with a viewpoint revisited: Descriptive norms and hotel guests’ towel reuse behavior. PLOS ONE, 9(8), e104086. https://doi.org/10.1371/journal.pone.0104086
  • Carrington, M. J., Neville, B. A., & Whitwell, G. J. (2010). Why ethical consumers don’t walk their talk: Towards a framework for understanding the gap between the ethical purchase intentions and actual buying behaviour of ethically minded consumers. Journal of Business Ethics, 97, 139–158. https://doi.org/10.1007/s10551-010-0501-6
  • Ertz, M., Karakas, F., & Sarigöllü, E. (2016). Exploring pro-environmental behaviors of consumers: An analysis of contextual factors, attitude, and behaviors. Journal of Business Research, 69(10), 3971–3980. https://doi.org/10.1016/j.jbusres.2016.06.010
  • Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A room with a viewpoint: Using social norms to motivate environmental conservation in hotels. Journal of Consumer Research, 35(3), 472–482. https://doi.org/10.1086/586910
  • Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54(7), 493–503. https://doi.org/10.1037/0003-066X.54.7.493
  • Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. https://doi.org/10.1016/S0065-2601(06)38002-1
  • Griskevicius, V., Tybur, J. M., & Van den Bergh, B. (2010). Going green to be seen: Status, reputation, and conspicuous conservation. Journal of Personality and Social Psychology, 98(3), 392–404. https://doi.org/10.1037/a0017346
  • Hainmueller, J., Hiscox, M. J., & Sequeira, S. (2015). Consumer demand for fair trade: Evidence from a multistore field experiment. Review of Economics and Statistics, 97(2), 242–256. https://doi.org/10.1162/REST_a_00467
  • Hardisty, D. J., Johnson, E. J., & Weber, E. U. (2010). A dirty word or a dirty world? Attribute framing, political affiliation, and query theory. Psychological Science, 21(1), 86–92. https://doi.org/10.1177/0956797609355572
  • Hines, J. M., Hungerford, H. R., & Tomera, A. N. (1987). Analysis and synthesis of research on responsible environmental behavior: A meta-analysis. The Journal of Environmental Education, 18(2), 1–8. https://doi.org/10.1080/00958964.1987.9943482
  • Johnstone, M.-L., & Tan, L. P. (2015). Exploring the gap between consumers’ green rhetoric and purchasing behaviour. Journal of Business Ethics, 132, 311–328. https://doi.org/10.1007/s10551-014-2316-3
  • Klöckner, C. A. (2013). A comprehensive model of the psychology of environmental behaviour—A meta-analysis. Global Environmental Change, 23(5), 1028–1038. https://doi.org/10.1016/j.gloenvcha.2013.05.014
  • Kollmuss, A., & Agyeman, J. (2002). Mind the gap: Why do people act environmentally and what are the barriers to pro-environmental behavior? Environmental Education Research, 8(3), 239–260. https://doi.org/10.1080/13504620220145401
  • Lange, F., & Dewitte, S. (2019). Measuring pro-environmental behavior: Review and recommendations. Journal of Environmental Psychology, 63, 92–100. https://doi.org/10.1016/j.jenvp.2019.04.009
  • Luchs, M. G., Naylor, R. W., Irwin, J. R., & Raghunathan, R. (2010). The sustainability liability: Potential negative effects of ethicality on product preference. Journal of Marketing, 74(5), 18–31. https://doi.org/10.1509/jmkg.74.5.018
  • Mertens, S., Herberz, M., Hahnel, U. J. J., & Brosch, T. (2022). The effectiveness of nudging: A meta-analysis of choice architecture interventions across behavioral domains. Proceedings of the National Academy of Sciences, 119(1), e2107346118. https://doi.org/10.1073/pnas.2107346118
  • Nisa, C. F., Bélanger, J. J., Schumpe, B. M., & Faller, D. G. (2019). Meta-analysis of randomised controlled trials testing behavioural interventions to promote household action on climate change. Nature Communications, 10, 4545. https://doi.org/10.1038/s41467-019-12457-2
  • Osbaldiston, R., & Schott, J. P. (2012). Environmental sustainability and behavioral science: Meta-analysis of proenvironmental behavior experiments. Environment and Behavior, 44(2), 257–299. https://doi.org/10.1177/0013916511402673
  • Reisch, L. A., & Zhao, M. (2017). Behavioural economics, consumer behaviour, and consumer policy: State of the art. Behavioural Public Policy, 1(2), 190–206. https://doi.org/10.1017/bpp.2017.1
  • Sheeran, P., & Webb, T. L. (2016). The intention–behavior gap. Social and Personality Psychology Compass, 10(9), 503–518. https://doi.org/10.1111/spc3.12265
  • Steg, L., & Vlek, C. (2009). Encouraging pro-environmental behaviour: An integrative review and research agenda. Journal of Environmental Psychology, 29(3), 309–317. https://doi.org/10.1016/j.jenvp.2008.10.004
  • Stern, P. C. (2000). New environmental theories: Toward a coherent theory of environmentally significant behavior. Journal of Social Issues, 56(3), 407–424. https://doi.org/10.1111/0022-4537.00175
  • Vermeir, I., & Verbeke, W. (2006). Sustainable food consumption: Exploring the consumer “attitude–behavioral intention” gap. Journal of Agricultural and Environmental Ethics, 19, 169–194. https://doi.org/10.1007/s10806-005-5485-3
  • White, K., Habib, R., & Hardisty, D. J. (2019). How to SHIFT consumer behaviors to be more sustainable: A literature review and guiding framework. Journal of Marketing, 83(3), 22–49. https://doi.org/10.1177/0022242919825649
  • Young, W., Hwang, K., McDonald, S., & Oates, C. J. (2010). Sustainable consumption: Green consumer behaviour when purchasing products. Sustainable Development, 18(1), 20–31. https://doi.org/10.1002/sd.394