Interactive effects of stimulus probability and significance on the skin conductance response

This 1982 study by Ben-Shakhar, Lieblich, and Kugelmass investigated how stimulus significance and stimulus probability jointly influence skin conductance responses (SCRs) in a Concealed Information Test paradigm. Using a card-selection task with 140 participants divided into informed and uninformed groups, the study found not only expected main effects of relevance and probability on SCRs, but a significant interaction showing that the probability-driven enhancement of SCRs is amplified when stimuli carry personal significance.

Laboratory analog experiment with 140 participants (80 informed, 60 uninformed) who were exposed to verbally presented number-based stimulus series with manipulated relative frequencies, while skin conductance responses were recorded as the primary psychophysiological measure.

This Hebrew University study investigated the fundamental psychophysiological mechanisms underlying the Concealed Information Test by examining how stimulus probability and significance interact to produce skin conductance responses. Using an elegant experimental design with informed and uninformed participant groups, researchers systematically varied the frequency of personally relevant number stimuli while recording electrodermal activity. The key finding was a significant interaction between significance and probability - these factors amplified each other's effects rather than combining additively. This demonstrated that probability-based orienting responses are substantially enhanced when stimuli carry personal meaning, providing crucial theoretical support for understanding why the CIT effectively detects concealed information.

These findings provide important theoretical validation for the CIT's effectiveness by demonstrating the psychophysiological mechanisms that enhance responses to concealed information. The interaction effect helps explain why guilty suspects show heightened responses to crime-relevant details and informs optimization of stimulus presentation protocols in polygraph examinations.

Background & Context

The psychophysiology of deception detection rests on a fundamental question: why does the electrodermal system respond differently to personally meaningful stimuli compared to neutral ones? By the early 1980s, researchers had established that the skin conductance response (SCR) — a measure of sweat gland activity reflecting autonomic arousal — was sensitive to both the novelty and the significance of a stimulus. What remained poorly understood was how these two properties interact when they co-occur in an experimental setting.

Ben-Shakhar, Lieblich, and Kugelmass published this investigation in Psychophysiology in 1982, addressing precisely this gap. Their work emerged from a broader Hebrew University research program examining the psychophysiological underpinnings of the Concealed Information Test (CIT), also known as the Guilty Knowledge Test. The CIT relies on the premise that a guilty subject will show heightened autonomic responses to crime-relevant stimuli embedded among innocuous alternatives — and understanding what drives those responses is critical to the test's validity.

Prior to this study, probability and significance were often treated as independent variables in orienting response research. The key theoretical question — whether they interact multiplicatively or additively — had direct implications for how examinees process relevant items during a polygraph examination and why detection works at all.

Research Design & Methodology

The study dealt with the effects of stimulus significance and stimulus probability on SCRs. Eighty subjects in an "informed" group chose a numbered card, while 60 subjects in an "uninformed" group chose a blank card. All subjects were then presented with a series of stimuli in the form of questions about numbers, and the relative frequencies of the relevant (chosen) numbers were systematically manipulated.

This design closely mirrors real-world CIT conditions. Key methodological features included:

• Two participant groups: an "informed" group (n = 80) with genuine personal significance attached to a specific number, and an "uninformed" control group (n = 60) with no such significance
• Stimulus frequency manipulation: the relative frequency (probability) of the target number was varied across conditions to assess the independent and interactive effects of probability
• Psychophysiological measure: skin conductance responses (SCRs) as the primary dependent variable, recorded under standardized laboratory conditions
• Between-groups comparison: allows isolation of significance-driven SCR enhancement from general novelty/probability-driven responses
• Stimulus format: verbal questions about numbers, maintaining ecological relevance to standard polygraph examination formats

The experimental logic was elegant: by crossing stimulus significance (informed vs. uninformed) with stimulus probability (varying relative frequency), the team could determine whether these factors combine in a simple additive fashion or whether significance amplifies the probability-based orienting response above and beyond what either factor alone would predict.

Results & Key Findings

The headline finding was a significant interaction between stimulus significance and stimulus probability — the two factors did not act independently but amplified each other's effects on the SCR.

The results showed expected main effects of both stimulus relevance and stimulus probability on SCRs. More critically, there was a significant interaction: the probability effect (i.e., greater responsivity to the less frequent stimulus) was substantially stronger when the series of stimuli was comprised of relevant and neutral stimuli, compared to when two different neutral stimuli were used.

Key findings from the study include:

• Main effect of significance: Informed subjects showed reliably larger SCRs to their chosen (relevant) number than uninformed subjects showed to any number
• Main effect of probability: Lower-frequency stimuli elicited larger SCRs in both groups, consistent with classic orienting response theory
• Interaction effect: The probability-driven SCR enhancement was significantly greater in the presence of personally significant stimuli — the two factors combined super-additively
• Control condition baseline: When both stimuli were neutral (uninformed group), frequency manipulations still produced SCR differences, but of a smaller magnitude
• Implication for the CIT: Guilty subjects benefit from a "double boost" in SCR — from both the personal relevance of the crime item and from its relative infrequency in the stimulus series

The interaction pattern was theoretically important because it showed that stimulus significance does not merely add a fixed increment to SCR magnitude — it fundamentally alters the way the autonomic nervous system weights probabilistic information about incoming stimuli.

Discussion & Significance

This study made a significant theoretical contribution by demonstrating that the orienting response in the CIT context is not simply a function of novelty or significance in isolation, but emerges from their dynamic interplay. The findings support a model in which personal relevance modulates the system's sensitivity to statistical regularities in the stimulus environment — a finding that aligns with later cognitive models of attention and predictive processing.

For polygraph science, the practical consequence is meaningful: the CIT is not merely detecting "surprise" at an unexpected stimulus, nor is it simply detecting emotional significance. Three experiments examining the effects of stimulus novelty and significance on the skin conductance component of the orienting response by the same research group later elaborated on these mechanisms, suggesting the 1982 study launched a productive line of inquiry into the multi-factorial nature of electrodermal responses in deception-detection contexts.

The Hebrew University group, including Lieblich, Kugelmass, and Ben-Shakhar, had earlier demonstrated the efficiency of GSR detection of information as a function of stimulus set size, and the 1982 study complemented that work by clarifying why those set-size effects exist: larger sets of alternatives reduce the relative frequency of any single item, thereby enhancing the probability-significance interaction that drives detection. This helped unify two previously separate threads of CIT research.

Limitations & Considerations

Several methodological considerations temper the interpretation of these findings:

• Laboratory analog setting: The card-selection paradigm, while well-controlled, lacks the high-stakes emotional context of a real criminal investigation — stakes which could modulate both the significance effect and the interaction
• Single physiological channel: Reliance solely on SCRs excludes other polygraph channels (respiration, blood pressure, heart rate) that may respond differently to these factorial manipulations
• Verbal stimulus format: Questions about numbers may not generalize perfectly to the more complex, emotionally charged stimuli used in actual forensic CIT examinations
• Uninformed group as control: While methodologically clean, the uninformed group cannot fully replicate the psychological state of an innocent suspect who has been informed about the crime but has no personal knowledge
• Historical instrumentation: 1982-era electrodermal recording technology and analog scoring methods are less precise than modern computerized systems, potentially introducing measurement noise

Additionally, the study's relatively brief publication length (pages 112–114 in Psychophysiology) suggests a concise report format, which may mean that some procedural details, effect sizes, and exploratory analyses were not fully elaborated in print — a common constraint of journal reporting norms of that era.

Practical Applications

For polygraph examiners, this study offers important guidance on test construction in the Concealed Information Test. The interaction between probability and significance means that the ratio of relevant to neutral stimuli within a question series is not a trivial design choice — it directly affects the magnitude of the SCR differential that forms the basis of detection. Examiners who use a standard series of five or six alternatives are, in effect, exploiting this interaction by keeping the probability of any single item low.

For consumers and legal professionals evaluating the CIT's scientific basis, this research helps explain why the test works at a mechanistic level. The SCR advantage for crime-relevant stimuli is not purely emotional or motivational — it is partly driven by an automatic, probability-sensitive orienting system that responds more strongly to statistically infrequent items when those items also carry personal meaning. This dual-mechanism foundation strengthens the theoretical legitimacy of CIT-based psychophysiological assessment and supports its continued use in appropriate forensic and investigative contexts.