Response conflict, sensory details and other theory-based approaches to memory and deception detection
Different procedures have been developed that aim at detecting deception, i.e. the “deliberate attempt, without forewarning, to create in another a belief which the communicator considers to be untrue” (Vrij, 2008, p. 15). The theoretical underpinnings of these procedures are divers and differ in precision. This paper will propose a framework for deception research based on established psychological theories. Also, these theories will be linked to relevant neuropsychological and psychophysiological findings, while taking into consideration the typical restraints given in the context of applied deception detection. In doing so it seems reasonable to break down the complex process of deception versus truthful memory-based reporting into the simpler processes that can be theoretically expected to be involved.
In the following, I will begin with a brief review of the most important psychophysiological procedures of deception and memory detection and its theoretical foundations. Then, the instance of mental response conflict, a psychological mechanism often assumed to occur in deception but not in truthful accounts, shall illustrate the proposed approach and its potential impact. In the end, I will indicate other mental processes whose involvement in deceiving would be interesting to investigate from a theoretical-analytical and, building on that, empirical perspective, potentially using physiological, behavioral and verbal indicators.
Early psychophysiological approaches were based on the assumption that anxiety in the situation of an examination discriminates between deceptive and truthful testimonies (Larson, 1932). However, this idea seems little plausible in most forensic scenarios (Vrij & Ganis, 2013). Other approaches to the detection of deception and memory rely on the processes of heightened cognitive load or recognition. For instance, the Concealed Information Test (CIT; Lykken, 1959) does not measure deception as such but can help identify it. It is theoretically based on the orienting response (Vrij & Ganis, 2013). During a CIT the person examined is confronted with objects or facts that are only known by those involved in the event of interest. A psychophysiological reaction to these stimuli that characteristically deviates from the reactions to control stimuli in terms of electrodermal activity, heart rate and amplitude of the P300 EEG-component is a valid indicator of recognition of the stimulus or piece of information (Meijer, Selle, Elber, & Ben-Shakhar, 2014, Verschuere & Meijer, 2014). In the laboratory, concealed information has been shown to be detectable with the CIT with an overall accuracy above 80% (Verschuere & Meijer, 2014). However, the CIT is useful only in cases in which stimuli that will be recognized by those directly involved in a (past) crime (and by nobody else) are available to the investigators. Therefore, the procedure is usually not applicable to crimes that happened a long time ago, that are planned for the future, or whose details are either publicly known or not known at all.
Hence, other approaches have concentrated on the detection of outright lies as opposed to information concealing. For instance, the Comparison Question Tests (CQT; Raskin & Kircher, 2013) measure the interviewee’s physiological responses (e.g. electrodermal activity, blood pressure, respiration and others) when answering crime-relevant questions (RQ) and when lying on matters that are not directly relevant (comparison questions; CQ). A stronger physiological response to the RQ than to the CQ is interpreted as a sign of lying. The nature of the mechanism that mediates the stronger psychophysiological reaction associated with lying in the CQT is not clear (Johnson, 2013; National Research Council, 2003, S. 213; Vrij & Ganis, 2013, 2013). An involved process may be an increase in cognitive load in liars, caused by anxious thoughts, conflicting response tendencies and (other) monitoring processes (Vrij & Ganis, 2013). Cognitive load theory (Sweller, 1988), originally formulated in the context of cognitive schema acquisition, is also the main theoretical basis of deception detection research using EEG and fMRI. The data so far indicate that when lying, response latencies are increased and brain areas associated with working memory, conflict between different response tendencies and response inhibition show heightened activation (Johnson, 2013; Greene & Paxton, 2009; Johnson, Barnhardt, & Zhu, 2005). Among other possible theoretical implications, these results from examinations in laboratory setting support the idea that lying is qualitatively different from truth-telling in that it necessitates the activation of two conflicting (re-)action tendencies: a mental representation of something one believes to be true or real and a deceptive alternative, the two tendencies being contradictory. Truthful reports, on the other hand, would require memory retrieval only. This idea is at the core of what Davis (1961) calls the conflict theory of deception. Although Davis mentions the psychoanalytic origin of conflict theory and names “habit” (p. 161) as the reason why the truthful response impulse emerges, his explanation of response conflict remains unsatisfying. Actually, even today (e.g. Vrij & Ganis, 2013) the conflict theory of deception does not have explanatory or predictive power beyond the intuitive hypothesis that truthful and deceptive responses mentally conflict in liars. However, while the production of a deceptive response is logically necessary for performing a lie, the activation of a truthful response is not. It is at least conceivable that a knowingly wrong statement is made without simultaneously representing a corresponding truthful alternative in working memory. Nevertheless, the idea that an automatically activated truthful response, Johnson (2013) calls it the pre-potent truthful response, occurs whenever a subject is confronted with a question seems to be generally accepted (Debey, Ridderinkhof, Houwer, Schryver, & Verschuere, 2015; Hacker, Kuhlman, & Kircher, 2013; Vrij, Granhag, Mann, & Leal, 2011). Therefore, it seems desirable to adjust conflict theory to related psychological theories that are more precise and empirically supported.
For instance, deception can be interpreted in terms of Wegner's (1994) ironic process theory. This theory proposes that in the attempt to control mental states, a monitoring process is installed in addition to the operating process. While the operating process works in the intended direction, the monitoring process searches for mental content that is inconsistent with the intention in order to evaluate the need for (an increase in effort in) the operating process. There is evidence for a neural circuit including the anterior cingulate cortex (ACC) and the dorsolateral prefrontal cortex (DLPFC) that seems to be responsible for such a monitoring process (Johnson, 2013). Ironically, according to Wegner (1994) the monitoring process tends to generate exactly the mental processes and contents that the whole controlling endeavor is meant to avoid. If the intention is to lie, for instance, the operating process would generate a deceptive response to a question. At the same time, the monitoring process would search for information that is inconsistent with this response because such information would indicate failure and the need to reinitiate the operating process or to increase effort. However, in searching for task-incongruent information, the monitoring process is likely to come upon the truthful response, i.e. a response based on memory concerning the event (or attitude or intention) in question. Ironic process theory further holds that in situations of particularly high cognitive and/or emotional load the operating process is weakened. Hence, the mental contents activated by the monitoring process, indicating failure of mental control, dominate conscious experience. In people intending to lie, increased stress and cognitive load should therefore result in either involuntary execution of the truthful responses or at least in an intensive conflict between truthful response and lie, resulting in high response latencies. Research showing less spontaneous lying in situations of high cognitive load (Mann, Vrij, Leal, Vernham, & Geven, 2015) supports this view. Theoretically, the described reasoning is in accordance with the general strategy to magnify cognitive load in order to improve distinguishability of liars and truth-tellers (Vrij et al., 2011; Vrij, Fisher, & Blank, 2015).