Complex cognitive functions require interactions within and between different brain regions by direct anatomical connections or synchronous activation. As such, damage to any region involved in a cognitive process has the potential to affect its function. Impulsivity is a multifaceted construct that, when dysfunctional, contributes to many psychiatric conditions. The striatum has been implicated as an integral part of the neural circuitry of impulsivity. The current work aims to contribute to the understanding of neural dysfunction underlying disorders of impulsivity by examining how striatal damage affects impulsive behavior. It also aims to improve our understanding of whether neural processes involved in impulsivity are also involved in maintaining awareness of one's thoughts and actions. No studies have systematically examined the extent to which damage to the striatum correlates with both changes in impulsive behavior and changes in self-awareness of impulsive personality. In the first experiment, I examined the effects of focal unilateral striatal damage on self-awareness of impulsivity and other personality traits. I predicted that participants with striatal damage (SD) would have less self-awareness of changes in impulsivity and other personality traits after brain damage, as compared to brain damage comparisons (BDC), due to indirect disruption of neural networks responsible for self-referential processing. I tested this prediction using self and collateral versions of the Barratt Impulsiveness scale (BIS) and the Iowa Scales of Personality Change. In partial support of my hypothesis, there were mean differences in self- and collateral-reported impulsivity on the BIS, with self ratings higher than collateral ratings in the SD group. There were no significant differences in the correlations between self- and collateral-reports for current impulsivity, change in impulsivity, or change in other personality traits. In the second experiment, I examined the effects of focal unilateral striatal damage on laboratory measures of impulsivity. I predicted that participants with striatal damage would exhibit lower levels impulsivity than brain damaged comparisons due to structural loss of regions involved in reward/motivation and motor activity. I tested this using impulsive action tasks (Go/NoGo and Stop Signal Tasks) and impulsive choice tasks (Delay and Probability Discounting). In contrast to my hypothesis, SD participants did not exhibit less impulsive action or impulsive choice than BDC participants. In the third experiment, I examined the effects of focal unilateral striatal damage on the integrity of frontostriatal resting state functional connectivity. I predicted that participants with striatal damage would exhibit alterations in functional connectivity between the remaining regions of the frontostriatal network. I tested this by comparing the strength of functional connectivity of the caudate head and ventromedial prefrontal cortex. While my hypothesis was not directly supported, the data showed interesting trends that warrant further exploration. These included stronger caudate-vmPFC resting state functional connectivity on the lesion side, and weaker functional connectivity on the non-lesioned side in striatal participants compared to brain damaged comparisons. Together, these experiments suggest that although unilateral striatal damage does not appear to affect subjective reports or laboratory measures of impulsivity, it may affect the underlying neural networks utilized by the striatum, as evidenced by changes in frontostriatal resting state functional connectivity. This work extends our understanding of the neurobiology of impulsive behavior and self-awareness, at systems level, and may help pave the way for treatments of those with brain injury, such as traumatic brain injury and stroke patients, or psychiatric disorders involving impulsivity.

Investigation of the neural correlates of the self has implicated a network of brain regions including the medial prefrontal cortex (MPFC), posterior cingulate (PCC), precuneus (pC), and inferior parietal lobule (IPL). At the same time, recent neuroimaging work has identified the Default Mode Network (DMN), a network of brain regions that are highly active at `rest' (without an active cognitive task). While the functional significance of the DMN remains unknown, converging evidence suggests that the DMN might be critical for self-referential processing (e.g., introspection). In this dissertation, I tested this hypothesis using a lesion approach. In the first experiment, I examined the critical role of the DMN hubs (MPFC, IPL) in autobiographical memory (AM) retrieval. I predicted that if the DMN hubs were critical for AM, then lesions to either the MPFC or IPL should result in AM retrieval impairments. I tested this prediction using the Iowa Autobiographical Memory Questionnaire (IAMQ), a questionnaire that assessed retrograde AM retrieval. In support of the prediction, lesions to the MPFC and IPL were associated with significant AM retrieval impairments. While not predicted, AM retrieval deficits were also associated with lesions in medial and lateral temporal cortices, regions also considered part of the DMN. In the second experiment, I tested the critical role of the DMN hubs in the self-reference effect (SRE), a well-known memory advantage conferred by self-related processing. I predicted that if the DMN hubs were critical for the SRE, then damage to the MPFC or IPL should diminish the effect. I used a standard personality trait judgment paradigm to test this prediction. In partial support of the prediction, I found that damage to the MPFC abolished the SRE with a "self" specific deficit. While IPL damage was associated with a diminished SRE, the effects were not significant. In the third experiment, I tested the hypothesis that the DMN is critical for accurate knowledge of one's personality. I predicted that if the DMN hubs are critical for accurate self-knowledge, then damage to either the MPFC or IPL should be associated with less accurate personality reports. In partial support of the prediction, MPFC and IPL groups demonstrated less accurate personality ratings. However, performance for all lesion groups was comparable and not significantly different from healthy subjects. In the fourth experiment, I sought to test the hypothesis that the DMN is critical for mind wandering (MW). I predicted that if the DMN hubs are critical for MW, then damage to the MPFC and IPL should result in decreased MW. To test this prediction, I used two approaches: 1) an experience sampling method (Sustained Attention to Response Task), and 2) a self-report measure (Imaginal Processes Inventory scale of MW). Contrary to my prediction, IPL lesions were associated with increased MW on the SART. By contrast, in support of the prediction, both MPFC and IPL lesions were associated with significant self-reported decreases in MW. Together, these experiments provide some evidence to support the hypothesis that the DMN is critical for self-referential processing. Future work might investigate the impact of DMN lesions on other self-processes (e.g., self-agency).

Empathy plays a crucial role in our relationships with others and enhances personal well-being. The brain areas that are critical for the experience of on-line empathy and empathic behavior are not known. The current study investigated the neural substrates of empathy through the examination of whether the ventromedial prefrontal cortex (VMPC) is critical for empathy. For the first time, on-line empathic experience and behavior were measured in patients with brain damage to the VMPC. Six patients with bilateral damage to the VMPC were case-matched on specific demographic and neuropsychological criteria to two comparison groups: a brain damage group and a healthy adult group. On-line empathy was induced in an ecologically-valid manner in which the participant experienced live the sorrow of another person. The participant thought they would be playing an economic game against two opponents. However, during the study the participant overheard their game opponent experience deep sadness, revealing that it was the anniversary of their son's death (empathic induction.) A comparison neutral induction involved the participant overhearing their opponent converse with the research assistant about a neutral topic. On-line empathic experience was measured by a questionnaire completed before and after the inductions. Empathic behavior was measured implicitly through an economic game. It was defined as the degree of behavioral change on the game as a result of the empathic induction (after accounting for baseline behavior.) The economic game used to measure empathic behavior was the Repeated Fixed Opponent variant of the well-validated Ultimatum Game. This particular variant had not been studied in participants of a similar age range to the patient sample (younger and older adults). Furthermore, there is evidence for some aging-related differences in behavior on economic games, providing additional rationale to examine the behavior of healthy younger and older adults on the game. Consequently, game behavior of younger and older adults was measured and then used to implement a model of healthy game behavior in the experiment that investigated empathy in patients with damage to the VMPC. Patients with damage to the VMPC experienced poor on-line empathy and showed poor empathic behavior. Patients with brain damage to the VMPC reported significantly less on-line empathy than patients with brain damage to other regions. Empathic behavior was not shown by patients with damage to the VMPC as a result of the empathic induction and their behavior was significantly different from both the healthy and the brain damage comparison groups which showed increased empathic behavior due to the empathic induction. A specific role for the VMPC region in empathy was demonstrated by the finding that patients with damage to this region had less on-line empathy and empathic behavior than patients with brain damage to other regions. This study showed that the VMPC region of the brain is critical for empathy. Further research is needed to elucidate whether patients with brain damage to the VMPC show decreased empathic behavior in all domains or whether it is specific to monetary decision-making.

Cigarette smoking is an addictive behavior. There are two learned emotional responses to smoking that may be particularly important for promoting addiction to smoking: the pleasure obtained from the airway sensory effects of smoking (airway sensory pleasure) and the urge to smoke that is elicited by environmental smoking cues (cue-induced urge). The ventromedial prefrontal cortex (VMPFC) has been implicated in a variety of learned emotional and motivational responses to drug-associated sensory cues. This project set out to address the role of the VMPFC and related areas in airway sensory pleasure and cue-induced urge, as well as the role of this region in promoting smoking behavior in real life, by examining the effects of focal lesions within the VMPFC and related areas in human cigarette smokers. It was found that lesions of the VMPFC itself were associated with a marked impairment of cue-induced urge in the laboratory, which was paralleled by a reported reduction in the difficulty of abstaining from smoking in real life. At the same time, VMPFC lesions led to a relative sparing of airway sensory pleasure in the laboratory, which was paralleled by no change in the enjoyment from smoking in real life. In addition, it was found that VMPFC lesions were not associated with changes in real-life measures of smoking dependence. It was found that lesions of the insula, a region that is functionally related to the VMPFC, were associated with an ability to quit smoking easily, immediately, without relapses and without a lasting urge to smoke. However, among patients with insula lesions who continued to smoke after lesion onset, there were no appreciable impairments of airway sensory pleasure or cue-induced urge. The results suggest that, while VMPFC lesions may disrupt cue-induced urges, they do not disrupt dependence upon smoking. This may be because VMPFC lesions spare more implicit motivational processes, such as "habits" and "incentive salience," that can drive smoking behavior in the absence of a conscious desire to smoke. The results also suggest that the insula functions in psychological processes that may contribute to the difficulty of quitting smoking.