Lie detection and language comprehension

People who can't understand words are better at picking up lies about emotions.

Abstract

People are usually no better than chance at detecting lies from a liar's demeanour^1,2, even when clues to deceit are evident from facial expression and tone of voice³. We suspected that people who are unable to understand words (aphasics) may be better at spotting liars, so we tested their performance as lie detectors. We found that aphasics were significantly better at detecting lies about emotion than people with no language impairment, suggesting that loss of language skills may be associated with a superior ability to detect the truth.

Main

We studied the lie-catching abilities of ten patients who could understand individual words but who suffered severe deficits in comprehending spoken sentences after damage to the left cerebral hemisphere (LH). Their performance was compared with that of ten patients with damage to the right cerebral hemisphere (RH), ten healthy controls (C) and 48 undergraduates from the Massachusetts Institute of Technology (UC). Subjects watched a videotape in which each of ten people was shown twice consecutively: once attempting to conceal powerful negative emotions and once honestly revealing positive emotions. The sequence of the two interviews was random. Behavioural measurement showed that the interviews differed in subtle facial expressions and in pitch changes in the voice⁴.

Aphasics were significantly more accurate than controls at detecting lies. The mean of the LH group (0.61; s.d. = 0.10) was higher than that of the RH, C or UC groups (0.44, 0.47 and 0.46 respectively with standard deviations of 0.11, 0.16 and 0.16). An analysis of variance of the three matched groups (LH, RH and C) found a significant difference among groups (F_2,26 = 4.33, P < 0.03). A planned contrast analysis comparing the LH group against the others was statistically significant (F_1,26 = 12.95, P < 0.002) and the residual variance in the main effect was not significant; t-tests comparing the means of all four groups against the value of 5 (the value obtained by chance) revealed that only the LH group scored better than chance (F = 9.00, P < 0.02).

We then compared the performance of all groups on items where the clues were in facial expression (3 items), in pitch changes in the voice (1 item) or in the face and voice (6 items). LH patients do better when clues are in facial expression but not when the cues are from the voice alone ( Table 1).

Table 1 Success in interpreting lying cues

Our results support the untested claim that aphasic patients are unusually sensitive to deceitful behaviour^5,6. Perhaps damage to the circuitry underlying language comprehension results in the growth of compensatory skills in recognizing non-verbal behaviour. All but one of our aphasic patients were tested more than one year post-injury (the one who was tested within a year scored no better than chance). Although we cannot distinguish whether our patients were better lie detectors or simply better at detecting subtle cues to emotion, aphasics' abilities to recognize non-subtle facial expressions have, to our knowledge, never before been shown to be superior to those of controls^7,8. The superiority of aphasics to normal persons in any task is a rarity.