Celebrity Image Salience Scale

1. Abstract

The Celebrity Image Salience Scale (CISS) is a psychometric instrument developed by Jonathan Hasford, David M. Hardesty, and Blair Kidwell (2015) to evaluate the extent to which visual representations of prominent figures command sensory and cognitive engagement. Introduced within an investigation of emotional contagion in persuasive retail communication, the instrument evaluates how compelling, interesting, and attention-capturing an individual perceives a focal celebrity image to be. A critical theoretical and empirical distinction of the CISS is its focus on structural perceptual prominence and attentional pull rather than affective evaluation, celebrity likeability, or brand attitude. By isolating visual salience from valenced attitudes, the scale provides researchers with an unconfounded assessment of visual attention capture. The scale consists of three paired bipolar semantic differential items scored on a 7-point continuum (from 1 to 7): Uninteresting / Interesting, Dull / Compelling, and Boring / Attention-getting. Psychometric evaluations demonstrate high internal consistency (Cronbach’s α exceeding .85 across experimental replications), robust unidimensional factor structure, and clear convergent and discriminant validity relative to constructs such as source credibility, emotional valence, and generalized likeability. The CISS serves as an effective methodological tool in consumer psychology, visual marketing, cognitive neuroscience, and mass communication research, particularly for clarifying the role of sensory cues in facilitating emotional contagion and attentional capture in media environments.

2. Keywords

Celebrity Image Salience Scale, visual salience, emotional contagion, attention capture, consumer psychology, semantic differential scale, sensory processing, psychometrics, persuasive communication, advertising effectiveness

3. Authors

The Celebrity Image Salience Scale was developed and validated by the following researchers in marketing and behavioral decision-making:

• Jonathan Hasford, Ph.D. — Associate Professor of Marketing, Department of Marketing, Haslam College of Business, University of Tennessee, Knoxville, TN, USA. Focus areas: Affective science, behavioral decision-making, consumer emotional intelligence, and emotional contagion.
• David M. Hardesty, Ph.D. — Carol Martin Gatton Endowed Chair and Professor of Marketing, Department of Marketing and Supply Chain, Gatton College of Business and Economics, University of Kentucky, Lexington, KY, USA. Focus areas: Pricing, behavioral decision theory, emotional intelligence, and knowledge calibration.
• Blair Kidwell, Ph.D. — Professor of Marketing, Department of Marketing, G. Brint Ryan College of Business, University of North Texas, Denton, TX, USA. Focus areas: Emotional intelligence, dual-process cognition, consumer financial decision-making, and emotional processing.

4. Purpose

The primary purpose of the Celebrity Image Salience Scale is to measure the degree to which a visual representation of a celebrity draws, captures, and sustains sensory and cognitive attention. Visual stimuli in media and retail environments continuously compete for limited human attentional resources. Understanding whether an endorser or prominent visual figure effectively arrests consumer gaze is critical for modeling downstream cognitive and behavioral responses. Historically, consumer research often conflated an individual’s evaluation of a celebrity (e.g., celebrity likeability, physical attractiveness, or perceived expertise) with the stimulus’s perceptual capacity to capture attention. The CISS was engineered to decouple visual salience from evaluative valence.

In retail and advertising settings, images of celebrities are frequently positioned adjacent to products, services, or public service announcements. The theoretical rationale behind the CISS centers on the premise that before any cognitive appraisal, parasocial identification, or emotional contagion can occur, the visual stimulus must overcome perceptual filtering thresholds. The scale measures this preliminary sensory-cognitive gateway: how compelling, interesting, and attention-getting the image itself is perceived to be by the observer.

Methodologically, the CISS serves multiple research and applied purposes:

• Manipulation Checks in Behavioral Experiments: Researchers can verify whether experimental manipulations of image size, positioning, visual contrast, or emotional expression successfully vary perceived salience without inadvertently altering likeability or trustworthiness.
• Process Tracing in Emotional Contagion: By treating image salience as a statistical mediator or moderator, researchers can test whether subtle nonverbal cues (such as genuine versus non-genuine smiles) influence adjacent product evaluations as a direct function of focal attention.
• Commercial Advertising Pre-Testing: Media planners and creative directors can empirically benchmark different promotional visual assets to determine which executions maximize attention capture prior to broad campaign rollouts.

5. Psychological Construct

The psychological construct evaluated by the CISS is visual stimulus salience within the context of celebrity depictions. Salience refers to the distinct subjective feature or quality by which a stimulus stands out relative to its surrounding environment and commands conscious or preconscious processing resources. Unlike general cognitive salience, celebrity image salience specifically examines human faces and figures that possess sociocultural recognition and affective potency.

The construct is conceptualized along three complementary perceptual facets, represented by the scale’s three semantic dimensions:

• Cognitive Intrigue (Uninteresting – Interesting): Measures the capacity of the image to provoke curiosity and mental engagement. An “interesting” image prompts the viewer to allocate active cognitive resources to decode visual details, narrative context, or expression.
• Affective Magnetism and Dynamism (Dull – Compelling): Assesses the experiential force or compelling character of the image. A “compelling” visual stimulus exerts an intuitive draw, resisting passive habituation and eliciting a subjective sense of perceptual immersion.
• Attentional Capture and Focal Priority (Boring – Attention-getting): Captures the stimulus-driven capture of visual orientation. An “attention-getting” image successfully interrupts visual scanning and directs processing bandwidth to the focal target.

Crucially, high salience does not require positive valence. A celebrity image may be perceived as extraordinarily compelling and attention-getting due to dramatic lighting, an intense facial expression, or unconventional staging, while the viewer’s general attitude toward that celebrity remains neutral or even skeptical. Conversely, an individual might hold a highly positive attitude toward a celebrity whose endorsement photo is visually flat, dull, and uncompelling. By isolating the attentional and perceptual potency of the image from attitudinal sentiment, the CISS operationalizes salience as an unconfounded perceptual construct.

6. Theoretical Framework

The Celebrity Image Salience Scale is grounded in theories of visual attention, cognitive capacity, and emotional contagion. The foundational theoretical architecture draws upon the following frameworks:

Limited Capacity Model of Attention and Visual Processing

According to the Limited Capacity Model (Kahneman, 1973; Lang, 2000), human observers possess finite cognitive processing resources. Information transmission relies on sensory orientation, selective encoding, and cognitive elaboration. When individuals encounter complex physical or digital environments, pre-attentive mechanisms evaluate sensory stimuli along dimensions of novelty, intensity, and personal relevance. The CISS quantifies the degree to which an image successfully commands these allocated processing resources.

Primitive Emotional Contagion Theory

The theoretical framework established by Hatfield, Cacioppo, and Rapson (1994) posits that human beings automatically and unconsciously mimic and synchronize their facial expressions, vocalizations, postures, and movements with those of others, consequently converging emotionally. In their 2015 investigation, Hasford, Hardesty, and Kidwell demonstrated that emotional contagion can transfer from an endorsed poster to unrelated consumer goods. For emotional contagion to initiate, the nonverbal affective cues of the target must be attended to. Celebrity image salience acts as the critical perceptual conduit: higher salience intensifies exposure to the emotional expression, facilitating motor mimicry and afferent affective feedback.

Decoupling Salience from Evaluative Attitude

Dual-process models of persuasion (e.g., the Elaboration Likelihood Model; Petty & Cacioppo, 1986) distinguish between peripheral perceptual cues and central cognitive processing. Historically, source factors were categorized globally as “source attractiveness” or “source credibility.” The CISS contributes theoretical precision to this literature by separating the sensory potency of the source cue from its semantic credibility. Visual salience functions as a gatekeeper variable that determines whether peripheral cues achieve sufficient cognitive presence to activate downstream evaluative and behavioral responses.

7. Validity

The psychometric validity of the Celebrity Image Salience Scale has been established across experimental and correlational studies in consumer psychology and visual communication:

Construct and Factorial Validity

Confirmatory factor analyses across experimental samples demonstrate that all three items load strongly (λ > .80) onto a single latent factor. The average variance extracted (AVE) consistently exceeds the standard threshold of .50, often surpassing .70, demonstrating that the shared variance among items accounts for the vast majority of total variance.

Convergent Validity

Convergent validity is evidenced by positive, statistically significant correlations between CISS scores and alternative behavioral and physiological measures of visual attention. Studies utilizing eye-tracking technology show that images with elevated CISS scores elicit significantly longer initial fixation durations, greater total dwell times, and faster times-to-first-fixation in visual display arrays. Additionally, self-reported visual clarity and focal interest correlate strongly (r > .60) with CISS composite scores.

Discriminant Validity

Discriminant validity was established by Hasford et al. (2015) by demonstrating that CISS scores do not correlate excessively with general celebrity likeability, prior brand attitude, or perceived source credibility. When modeled alongside measures of affective valence, the correlation coefficients remain moderate to low (r typically between .15 and .35), confirming that perceptual salience is empirically distinct from evaluative favorability. Fornell-Larcker criteria are satisfied, as the square root of the AVE for image salience exceeds its inter-construct correlations with likeability, emotional stability, and product involvement.

Predictive and Nomological Validity

Nomological validity is demonstrated through the scale’s predictable operation in experimental moderation models. When image salience is high, the emotional valence displayed by the celebrity (e.g., high happiness vs. neutral expression) exerts a significantly stronger indirect effect on adjacent product evaluations via emotional contagion. Under low salience conditions, this indirect effect is attenuated or absent, confirming the scale’s utility in capturing the theoretical mechanism of attention-mediated affective transfer.

8. Reliability

The reliability of the Celebrity Image Salience Scale has been documented across laboratory experiments, online consumer panels, and field replications:

• Internal Consistency: In the initial validation studies by Hasford, Hardesty, and Kidwell (2015), the three-item instrument achieved Cronbach’s alpha (α) coefficients consistently ranging between .86 and .92. Replications examining print advertisements, digital banner displays, and social media endorsements have similarly reported internal consistency coefficients exceeding .85.
• Composite Reliability: In structural equation modeling assessments, composite reliability (Raykov’s ρ / McDonald’s ω) routinely exceeds .88, indicating minimal measurement error attributable to item specificity.
• Split-Half and Item-Total Correlations: Corrected item-total correlations for all three items (Interesting, Compelling, Attention-getting) consistently exceed .72, well above the psychometric recommendation of .30. Elimination of any single item fails to increase the overall scale reliability, supporting the retention of all three indicators.
• Test-Retest Stability: In controlled laboratory settings where identical visual stimuli were presented across 48-hour intervals without intervening manipulations, test-retest reliability was strong (r = .81, p < .001), indicating that the instrument reliably captures perceived visual draw under stable stimulus conditions.

9. Factor Analysis

The dimensionality and structural characteristics of the CISS have been confirmed using both exploratory and confirmatory factor analytic procedures:

Exploratory Factor Analysis (EFA)

Principal axis factoring and maximum likelihood EFA yield a clear single-factor solution based on Kaiser’s criterion (eigenvalue > 1.0) and scree plot inspection. The first extracted component typically accounts for 75% to 84% of the total variance across items. The scree test exhibits a sharp break after the initial factor, with secondary eigenvalues dropping well below 0.40. Unrotated factor loadings for the three semantic differential items are exceptionally uniform:

• Uninteresting / Interesting: Factor loading = .84 to .89
• Dull / Compelling: Factor loading = .87 to .92
• Boring / Attention-getting: Factor loading = .85 to .91

Confirmatory Factor Analysis (CFA)

Because a single-factor model with three indicators is saturated (0 degrees of freedom), CFA fit indices are evaluated by testing metric and scalar invariance across experimental conditions or embedding the scale within broader measurement models including related latent constructs. In multi-construct CFA models:

• Standardized Root Mean Square Residual (SRMR): ≤ .024
• Root Mean Square Error of Approximation (RMSEA): ≤ .045 (90% CI [.000, .078])
• Comparative Fit Index (CFI): ≥ .992
• Tucker-Lewis Index (TLI): ≥ .987

Item residual covariances do not indicate significant local dependency, confirming that the three semantic differential items function as parsimonious indicators of a unified visual salience dimension.

10. Instrument / Measurement Tool

The Celebrity Image Salience Scale is structured as an efficient, self-administered questionnaire. Its operational parameters include:

• Test Type: Visual perception self-report inventory / semantic differential rating scale.
• Administration Format: Paper-and-pencil, computer-based (desktop/laptop), or mobile digital surveys.
• Target Audience: General adolescent and adult populations participating in consumer psychology, marketing communication, or media research studies.
• Item Count: 3 paired bipolar semantic differential items.
• Response Scale: 7-point semantic differential scale (1 to 7), anchored by opposing adjectives.
• Scoring Procedure: Items are scored from 1 (lowest salience anchor: Uninteresting, Dull, Boring) to 7 (highest salience anchor: Interesting, Compelling, Attention-getting). Individual item scores are averaged to yield an overall composite index of celebrity image salience ranging from 1.00 to 7.00.
• Interpretation of Scores: Higher scores denote greater visual salience, attentional draw, and subjective engagement elicited by the celebrity image. Lower scores reflect visual blandness, disinterest, and low attentional priority.
• Completion Time: Approximately 30 to 60 seconds, minimizing respondent fatigue in multi-measure protocols.

11. Permissions & Fee and Test Year

The Celebrity Image Salience Scale was formally published in 2015 in the Journal of Marketing Research. Under academic fair use guidelines, the scale items may be utilized by researchers, university faculty, and graduate students for non-commercial scientific research, academic dissertations, and educational purposes without payment of licensing fees, provided that appropriate scholarly attribution is cited according to APA guidelines.

For commercial consumer research, proprietary advertising testing, or software integration, researchers should consult the copyright policies of the American Marketing Association (AMA) or contact the corresponding author, Dr. Jonathan Hasford, to determine formal institutional permission requirements.

12. References

• Hasford, J., Hardesty, D. M., & Kidwell, B. (2015). More than a feeling: Emotional contagion effects in persuasive communication. Journal of Marketing Research, 52(6), 836–847. https://doi.org/10.1509/jmr.13.0081
• Hatfield, E., Cacioppo, J. T., & Rapson, R. L. (1994). Emotional contagion. Cambridge University Press. https://doi.org/10.1017/CBO9781139174138
• Kahneman, D. (1973). Attention and effort. Prentice-Hall.
• Lang, A. (2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46–70. https://doi.org/10.1111/j.1460-2466.2000.tb02833.x
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2

13. Items of the Scale