In the present study, we tested our hypothesis that patients with Parkinson's disease have difficulty making deceptive responses due to dysfunction of the prefrontal cortex. As predicted, the patients could not successfully make deceptive responses compared with the healthy controls. Furthermore, consistent with previous neuroimaging studies with healthy individuals that have indicated an association between deception and the prefrontal cortex, FDG-PET imaging revealed that the patients’ failure in the deception task was significantly correlated with hypometabolism in the prefrontal cortex, regardless of age, sex and other possible confounding factors. To our knowledge, this is the first neuropsychological evidence that dysfunction of the prefrontal cortex is involved in the inability to inhibit true responses and produce deceptive responses in Parkinson's disease patients.
The results of the present study raise two important points. First, certain personality traits of Parkinson's disease patients (Menza, 2000
; Ishihara and Brayne, 2006
) might be at least partly explained by neuropsychological deficits. In other words, the cognitive deficits may have an influence on ostensible personality traits in Parkinson's disease patients. More specifically, the present results indicate that honesty in Parkinson's disease patients might result from impairment of the executive functions necessary for the processes involved in telling lies. Indeed, the patients showed worse performance in the verbal fluency task and the trail-making test (generally used as measures of executive function) compared with the normal controls. Although these tests are different from the deception task in terms of how the subjects respond (e.g. open-ended responses in verbal fluency and forced-choice responses in the deception task), and therefore are not likely to have direct impact on deception task performance, there is still a possibility that these tests partially share the cognitive and neural mechanisms of deception in terms of higher-order cognitive processes including executive function. In line with this idea, these task performances were significantly correlated with deception task performance. Future studies using an approach similar to that of the present study might further clarify the relationships between cognitive dysfunction and characteristic personality and behavioural traits in Parkinson's disease patients.
Second, the results reveal a direct association between a cognitive control system subserving deception and function of the prefrontal cortex. It is known that brain imaging of healthy people cannot provide direct evidence that a certain brain region is necessary for the performance of a specific cognitive task (Frackowiak et al
). That is, some activation in functional brain imaging studies may reflect brain activity that is not essential for the function of interest. Therefore, direct evidence is derived from loss-of-function studies. In the present study, we revealed that the right anterior prefrontal cortex and left dorsolateral prefrontal cortex, which have been activated during deception in a number of carefully designed imaging studies (for reviews, see Spence et al
; Sip et al
; Christ et al., in press
), are associated with making deceptive responses. In line with our results, a recent study using transcranial direct current stimulation provided evidence that manipulation of functions in the dorsolateral prefrontal cortex altered the speed and efficiency of deceptive responses (Priori et al
). Furthermore, the association between deception and the left dorsolateral prefrontal cortex in the present study is highly consistent with the findings of a series of neuroimaging studies that we have conducted with healthy individuals (Abe et al
Based on the previous findings and the present results, we propose that the left dorsolateral prefrontal cortex, the region implicated in a wide range of higher-level cognitive operations such as working memory (D’Esposito et al
; Salmon et al
) and resolution of response conflict (MacDonald et al
; Badre and Wagner, 2004
), plays a pivotal role in telling lies. The right anterior prefrontal cortex is also likely to play a critical role in integrating the multiple cognitive processes (Ramnani and Owen, 2004
) in deception. One might think that set-shifting deficits, one of the well-known cognitive deficits in Parkinson's disease (Ravizza and Ciranni, 2002
; Monchi et al
; Moustafa et al
; Nagano-Saito et al
), affect the results. However, our analysis of set-shifting effect on the response accuracy in Truth trials did not support this interpretation. We believe that our task does not simply measure set shifting, and that dysfunction of the left dorsolateral and right anterior prefrontal cortices specifically prevents Parkinson's disease patients from inhibiting true responses and producing deceptive responses.
It is important to determine how frontal executive dysfunction, possibly disrupting deceptive behaviour, is derived from the neuropathological changes observed in Parkinson's disease patients. One possibility is that prefrontal hypometabolism in Parkinson's disease patients results from degeneration of the substantia nigra pars compacta with subsequent depletion of dopamine in the striatum. A recent study suggests that the dorsolateral prefrontal circuit consisting of the dorsolateral prefrontal cortex, caudate nucleus, globus pallidus, substantia nigra, and thalamus (Cummings, 1993
; McPherson and Cummings, 2002
) is specifically associated with executive dysfunction in Parkinson's disease patients (Zgaljardic et al
). Alternatively, the executive dysfunction may reflect a functional disturbance of the frontal cortex itself caused by locally impaired mesocortical dopaminergic transmission (Mattay et al
). Although these two models are not mutually exclusive, there is controversy in the recent literature in that some researchers have argued that both the nigrostriatal and mesocortical pathways are disrupted in Parkinson's disease (Monchi et al
), whereas others have shown impaired nigrostriatal dopaminergic function with preserved mesocortical dopaminergic transmission in early Parkinson's disease (Sawamoto et al
). As for dopaminergic transmission, a study in which the ‘on’ and ‘off’ medication states are directly compared would also be useful. We can predict that dopaminergic medication would have a beneficial effect on the regions affected by depletion of dopamine, such as the caudate nucleus and thereby its connections to the dorsolateral prefrontal cortex, and that the ability to make deceptive responses would improve in Parkinson's disease patients. In fact, some previous studies have reported the beneficial effects of levodopa on cognitive performance, although it should be noted that the effects depend on the nature of the task (Gotham et al
; Cools et al
; Lewis et al
In conclusion, our results provide new evidence that damage to the prefrontal cortex disrupts the processes involved in making deceptive responses in Parkinson's disease patients. It appears that the ‘honesty’ of patients is caused by an impaired ability to deceive others that results from brain dysfunction caused by the disease. However, there are some limitations of the present study that should be borne in mind for future studies. First, the present study examined only the processes associated with executive control during deception. The participants were instructed to tell a lie, which cannot be viewed as being the same as deception in real life. The neural bases of genuine deception or immoral lying should be investigated further in both healthy individuals and brain-damaged patients. Second, it remains a possibility that the association between difficulty deceiving others and prefrontal dysfunction may not be specific to Parkinson's disease patients, and further studies are needed to examine whether patients with other neurological disorders affecting the prefrontal cortex show similar deficits (see Spence and Kaylor-Hughes, 2008
). Third, the present study investigated only patients with mild Parkinson's disease of short duration. Whether our claim is true of patients in general is an important issue to be pursued. Finally, it is also important to determine how (and when) the brain pathology derived from Parkinson's disease causes specific personality traits together with explicit cognitive deficits. A longitudinal assessment with detailed neuropsychological assessment and multimodal neuroimaging in Parkinson's disease patients is required.