Affective Response to the Ad (Anxiety) (ARTT)

1. Abstract

The Affective Response to the Ad (Anxiety) scale, frequently designated within consumer psychology as a specialized sub-index of the Affective Response to the Theme/Ad (ARTT) battery, is a focused, three-item self-report psychometric instrument designed to evaluate situational, state-level anxiety elicited by promotional communications. Introduced to the advertising and consumer research literature by Loraine Lau-Gesk and Joan Meyers-Levy (2009), the scale operationalizes message-induced anxiety through three distinct semantic descriptors: Anxious, Nervous, and Worried. Participants rate the intensity of their emotional response on a 7-point semantic differential continuum ranging from 1 (“Not at all”) to 7 (“Very much / Extremely”). Psychometrically, the instrument functions as a unidimensional reflective index characterized by exceptional internal consistency reliability (routinely demonstrating Cronbachu2019s alpha coefficients exceeding .85 across experimental replications) and robust convergent, discriminant, and predictive validity.

The theoretical architecture of the scale is anchored in cognitive appraisal theories of emotion and the resource matching hypothesis. Lau-Gesk and Meyers-Levy conceptualized anxiety not merely as a negatively valenced affective state, but crucially as an emotion characterized by high cognitive resource demands stemming from low situational certainty and external attribution of control. By isolating message-evoked anxiety from general negative affect or sadness, the scale permits researchers to evaluate how distinct emotional appraisals systematically alter cognitive capacity, information processing depth, and subsequent attitudinal compliance. This instrument has become an empirical standard in advertising research, public health messaging, fear-appeal evaluation, and consumer behavior investigations exploring emotional persuasion.

2. Keywords

Affective Response to the Ad, ARTT, Advertising Anxiety, Emotional Persuasion, Cognitive Appraisal Theory, Resource Matching Hypothesis, State Anxiety, Consumer Emotion, Negative Affective Appeals, Persuasion Research, Psychometrics, Advertising Effectiveness.

3. Authors

The Affective Response to the Ad (Anxiety) scale was formulated and validated within consumer behavior research by:

• Loraine Lau-Gesk, Ph.D. u2014 Professor of Marketing at the Paul Merage School of Business, University of California, Irvine. Dr. Lau-Gesk specializes in consumer emotion, emotional appeals, cross-cultural marketing, and the processing of mixed and negative affect in advertising contexts.
• Joan Meyers-Levy, Ph.D. u2014 Professor of Marketing and Chair of the Marketing Department at the Carlson School of Management, University of Minnesota. Dr. Meyers-Levy is an internationally renowned scholar in consumer information processing, cognitive psychology, gender differences in advertising reception, and sensory marketing.

Their foundational work establishing the psychometric properties of this scale in consumer contexts was published in the Journal of Consumer Research (2009), Volume 36, Issue 4, pages 585u2013599.

4. Purpose

The primary purpose of the Affective Response to the Ad (Anxiety) scale is to capture the precise degree of transient, message-induced anxiety experienced by an individual during or immediately following exposure to an advertising stimulus. While early consumer psychology literature frequently collapsed emotional reactions into broad, bipolar dimensions of valence (positive versus negative affect) or general autonomic arousal, such macro-level taxonomies fail to capture the functional and qualitative differences between distinct negative emotional statesu2014such as anxiety, sadness, anger, disgust, or guilt.

In applied and theoretical research, the scale serves several critical objectives:

• Disentangling Valence from Cognitive Demands: As demonstrated by Lau-Gesk and Meyers-Levy (2009), emotions sharing identical negative valence exert dramatically divergent consequences on cognitive processing. Anxiety inherently requires substantial cognitive processing resources because it signals uncertain, ambiguous threats that demand vigilance. The scale enables researchers to verify that experimental manipulations successfully activate state anxiety rather than alternate negative feeling states (e.g., sadness, which entails low resource demands due to perceived certainty of loss).
• Evaluating Health and Public Service Announcements (PSAs): Social marketers and public health agencies routinely employ fear appeals to discourage harmful behaviors (e.g., smoking, drunk driving, vaccine hesitancy). The scale quantifies the acute emotional tension elicited by these communications, allowing investigators to assess whether the induced anxiety falls within an optimal persuasive window or exceeds cognitive capacity, leading to defensive avoidance or reactance.
• Testing Resource Matching Frameworks: The resource matching hypothesis posits that persuasion is maximized when the cognitive resources made available by a consumer match the cognitive resources demanded by the advertising message. By measuring felt anxiety, researchers can predict whether an audience will possess the spare working memory capacity necessary to elaborate on complex message arguments.
• Diagnostic Pretesting in Commercial Advertising: Commercial practitioners use the scale during copy testing to detect unintentional negative affect. For products offering comfort, security, or financial stability, inadvertent induction of anxiety can severely depress brand evaluation unless the advertised offering is positioned as an immediate, efficacious anxiety-reducing solution.

5. Psychological Construct

The construct measured by this instrument is Message-Evoked State Anxiety, a situationally activated affective and physiological reaction elicited specifically by exposure to promotional stimuli. Unlike trait anxiety, which represents a stable, enduring personality disposition toward perceiving environmental stimuli as threatening (as operationalized by Spielberger’s State-Trait Anxiety Inventory), message-evoked anxiety is episodic, acute, and directly attributed to the content, framing, imagery, or narrative arc of the commercial communication.

Core Dimensional Facets

The construct is comprised of three interrelated emotional markers captured by the scale’s items:

1. Anxious (Affective Tension & Apprehension): Represents the phenomenological experience of dread or unease concerning an impending, uncertain negative outcome. In an advertising context, this manifests when a message highlights prospective physical, financial, or social vulnerabilities that the consumer may face.
2. Nervous (Autonomic & Somatosensory Arousal): Captures the heightened psychophysiological agitation and jitteriness associated with sympathetic nervous system activation. This component reflects the somatic sensations of vigilance, restless alertness, and heightened readiness to respond to an environmental challenge presented in the advertisement.
3. Worried (Cognitive Intrusiveness & Rumination): Encompasses the repetitive, verbal-linguistic thoughts focused on negative possibilities and uncontrollable outcomes. Cognitive worry directly competes for central executive resources within working memory, depleting the cognitive bandwidth available to process ancillary arguments within the advertisement.

Crucially, within cognitive appraisal theory, anxiety is distinguished from other emotional constructs along two central appraisal dimensions: situational certainty and agency/control. While sadness is characterized by high certainty that an unfortunate loss has already transpired, anxiety is defined by low certainty (high ambiguity) about an impending threat and low perceived individual control. This cognitive ambiguity triggers an urgent need to dedicate attentional resources toward resolving uncertainty, thereby imposing heavy resource demands on the human cognitive architecture.

6. Theoretical Framework

The Affective Response to the Ad (Anxiety) scale is theoretically underpinned by the integration of two major paradigms in psychology and consumer research: Cognitive Appraisal Theory and the Resource Matching Hypothesis.

Cognitive Appraisal Theory

Pioneered by Richard Lazarus (1991) and extended to consumer psychology by Roseman (1991) and Smith and Ellsworth (1985), cognitive appraisal theory posits that emotions are not direct, automatic responses to environmental events, but rather the consequence of cognitive evaluations along specific appraisal dimensions. These dimensions include:

• Motivational Valence: Whether the event is perceived as positive/congruent or negative/incongruent with personal goals.
• Certainty: The degree to which outcomes are predictable and known versus ambiguous and unpredictable.
• Attribution of Agency: Whether the event is caused by the self, another person, or impersonal situational forces.

Lau-Gesk and Meyers-Levy (2009) synthesized this literature to demonstrate that while sadness and anxiety both register as negatively valenced, they occupy polar opposites along the certainty dimension. Anxiety arises when an event is appraised as having potential negative consequences under conditions of uncertainty. Because the human brain evolved to prioritize survival threats, uncertain threats force an intense cognitive appraisal process: the individual must continuously monitor the environment, search memory for mitigation strategies, and evaluate coping mechanisms. Consequently, anxiety is theoretically characterized by exceptionally high cognitive resource consumption.

The Resource Matching Hypothesis

Formulated in consumer information processing by Meyers-Levy and Peracchio (1995), the resource matching hypothesis posits that persuasive impact depends on the alignment between the cognitive resources allocated by the consumer ($RA$) and the cognitive resources demanded by the message ($RD$). Optimal persuasion, recall, and attitudinal elaboration occur when $RA = RD$.

• If $RA < RD$, the consumer suffers cognitive overload, failing to process the ad’s substantive claims and resulting in superficial or frustrated evaluation.
• If $RA > RD$, the consumer possesses excess cognitive resources that are frequently redirected toward generating idiosyncratic counterarguments, cognitive distraction, or nitpicking the ad execution.

When an advertisement evokes anxiety, the emotional reaction itself consumes a substantial portion of working memory resources. Thus, the resource demands of the emotion ($RD_{emotion}$) must be added to the resource demands of the message arguments ($RD_{message}$). If an ad evokes high anxiety and simultaneously presents complex, multi-attribute product claims, total resource demands exceed available cognitive capacity ($RD_{total} > RA$), leading to compromised persuasion. The ARTT Anxiety scale provides the precise empirical parameter needed to quantify the emotional resource demand component within this theoretical model.

7. Validity

The psychometric validity of the Affective Response to the Ad (Anxiety) scale has been substantiated through rigorous experimental and empirical evaluations in consumer research.

Construct and Convergent Validity

Construct validity was demonstrated by Lau-Gesk and Meyers-Levy (2009) across multiple laboratory experiments. In manipulation check procedures, stimuli crafted to induce anxiety (e.g., ads depicting uncertain health risks, unpredictability, and imminent personal threat) generated statistically significant elevations on the composite anxiety index compared to neutral baseline advertisements and ads designed to induce sadness ($p < .001$). The three items (Anxious, Nervous, Worried) consistently exhibit strong inter-item correlations (typically $r = .72$ to $r = .86$) and load onto a single unrotated latent factor with standardized factor loadings routinely exceeding .80, indicating robust convergent validity at the latent construct level.

Discriminant Validity

Crucially, the scale exhibits distinct discriminant validity against other affective dimensions, most notably sadness, general negative affect, and anger:

• When modeled in confirmatory factor analysis alongside sadness markers (e.g., Sad, Depressed, Gloomy), a two-factor model demonstrates superior fit over a collapsed single-factor negative affect model ($\Delta \chi^2$ tests consistently significant at $p < .001$).
• Average Variance Extracted (AVE) for the anxiety construct consistently exceeds .70, comfortably surpassing the squared correlation ($\Phi^2$) between the anxiety factor and adjacent emotional factors (such as guilt or anger), thereby satisfying the Fornell-Larcker criterion.

Predictive and Criterion Validity

The predictive validity of the scale is highlighted by its ability to moderate and mediate downstream consumer responses in accordance with theoretical predictions:

• Cognitive Resource Depletion: Elevated scores on the anxiety scale reliably predict reduced recall of complex secondary ad copy, confirming that state anxiety drains central executive resources.
• Interaction with Message Complexity: As observed in experimental trials, when ARTT anxiety scores are high, consumers express more favorable attitudes toward ads that present simple, easy-to-process solutions, whereas complex message executions lead to attenuated brand attitudes due to cognitive overload.

8. Reliability

The three-item Affective Response to the Ad (Anxiety) instrument demonstrates exceptional internal consistency across diverse experimental samples, ad executions, and demographic cohorts.

Internal Consistency Metrics

• Cronbachu2019s Alpha ($lpha$): In the seminal investigation by Lau-Gesk and Meyers-Levy (2009), the composite reliability of the three-item index was exceptionally high, reporting an alpha coefficient of $lpha = .91$ in primary experimental conditions and $lpha = .89$ in replication studies. Subsequent marketing studies adopting the scale have consistently documented alpha coefficients ranging between $.86$ and $.94$.
• Composite Reliability (CR): In structural equation modeling (SEM) contexts, composite reliability values routinely exceed $.90$, far surpassing the established psychometric benchmark of $.70$.
• Inter-Item Correlations: Inter-item correlation matrices consistently exhibit values between $.68$ and $.84$, indicating that each semantic descriptor contributes substantial shared variance without redundancy.

Test-Retest Considerations

Because the instrument is explicitly designed as a state measure of immediate, transient affective reactions to a specific communicative stimulus, traditional longitudinal test-retest reliability across weeks or months is theoretically inappropriate. However, immediate post-exposure and delayed post-manipulation checks within the same experimental session demonstrate high consistency when the stimulus remains top-of-mind, verifying that the instrument captures a stable situational state rather than transient measurement error.

9. Factor Analysis

Structural evaluations of the Affective Response to the Ad (Anxiety) scale support a strict unidimensional factor structure.

Exploratory Factor Analysis (EFA)

When the three anxiety items are subjected to exploratory factor analysis (Principal Axis Factoring or Maximum Likelihood estimation) alongside other affective descriptors:

• A single dominant eigenvalue for the three items emerges (typically $lambda > 2.45$), accounting for over $80%$ of the total item variance.
• Item loadings onto the primary factor are exceptionally high: Anxious ($lambda pprox .88 – .94$), Nervous ($lambda pprox .82 – .89$), and Worried ($lambda pprox .85 – .92$).
• Cross-loadings onto orthogonal affective dimensions (such as sadness or positive cheerfulness) remain minimal (consistently below $.20$).

Confirmatory Factor Analysis (CFA)

When evaluated within a multi-construct affective measurement battery via Confirmatory Factor Analysis, the three-item measurement model yields excellent goodness-of-fit indices:

• Comparative Fit Index (CFI): $ge .99$
• Tucker-Lewis Index (TLI): $ge .98$
• Root Mean Square Error of Approximation (RMSEA): $le .045$ (with $90%$ confidence intervals encompassing zero)
• Standardized Root Mean Square Residual (SRMR): $le .020$

Because a three-item single-factor model is just-identified ($df = 0$), structural fit is formally assessed within broader models containing competing emotional constructs (e.g., sadness, guilt, positive affect). In these multi-factor structural models, the three anxiety items consistently demonstrate high standardized parameter estimates and minimal error covariance, confirming the empirical parsimony and structural integrity of the construct.

10. Instrument / Measurement Tool

• Construct Measured: Situational / Message-Evoked State Anxiety in response to an advertising stimulus.
• Scale Developer(s): Loraine Lau-Gesk and Joan Meyers-Levy (2009).
• Instrument Type: Self-administered paper-and-pencil or computerized questionnaire.
• Target Population: Adult consumers, student research pools, and survey panel participants evaluating promotional communications.
• Number of Items: 3 items.
• Administration Time: Under 1 minute (approximately 15 to 30 seconds).
• Response Format: 7-point semantic differential scale (1 = Not at all to 7 = Very much / Extremely).
• Item Descriptors:
  • Anxious
  • Nervous
  • Worried
• Scoring and Aggregation:
  • All items are positively keyed; there are no reverse-scored items.
  • The overall felt anxiety score is computed by calculating the arithmetic mean of the three responses:
  • $$\text{Anxiety Index} = \frac{\text{Anxious} + \text{Nervous} + \text{Worried}}{3}$$
  • Higher mean values denote greater intensity of message-evoked anxiety.

11. Permissions & Fee and Test Year

The Affective Response to the Ad (Anxiety) scale was published in its standardized empirical format in 2009 in the Journal of Consumer Research. Under established academic fair-use conventions, the three items may be freely reproduced and administered for non-commercial scholarly, educational, and scientific research without financial remuneration, provided appropriate citation is granted to the original authors (Lau-Gesk & Meyers-Levy, 2009).

Commercial marketing research firms, advertising copy-testing agencies, and corporate practitioners seeking to integrate the instrument into proprietary commercial diagnostic suites should consult the copyright policies of Oxford University Press / The Journal of Consumer Research, Inc., or seek direct permission from the copyright holders.

12. References

• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39u201350. https://doi.org/10.1177/002224378101800104
• Lau-Gesk, L., & Meyers-Levy, J. (2009). Emotional persuasion: When the valence versus the resource demands of emotions influence consumersu2019 attitudes. Journal of Consumer Research, 36(4), 585u2013599. https://doi.org/10.1086/599047
• Lazarus, R. S. (1991). Emotion and Adaptation. Oxford University Press.
• Meyers-Levy, J., & Peracchio, L. A. (1995). Understanding the effects of resource demands on cognitive processing: A visual illustration. Journal of Consumer Research, 22(3), 278u2013289. https://doi.org/10.1086/209451
• Roseman, I. J. (1991). Appraisal determinants of discrete emotions. Cognition & Emotion, 5(3), 161u2013200. https://doi.org/10.1080/02699939108411034
• Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 48(4), 813u2013838. https://doi.org/10.1037/0022-3514.48.4.813
• Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the State-Trait Anxiety Inventory. Consulting Psychologists Press.

13. Items of the Scale