## Abstract

We investigated sex-related differences in task performance and brain activity in the orbitofrontal cortex (OFC) and dorsolateral prefrontal cortex (DLPFC) during performance of a decision-making task (the Iowa Gambling Task). When men and women were examined separately, men activated extensive regions of the right lateral OFC and right DLPFC, as well as the left lateral OFC. In contrast, women activated the left medial OFC. Examining sex differences directly, men showed better task performance and greater lateralized brain activity to the right hemisphere than women. This was exemplified by greater activation in a large area of the right lateral OFC of men during their performance of the Iowa Gambling Task. In contrast, women had greater activation in the left DLPFC, left medial frontal gyrus and temporal lobe during this task. Thus, brain mechanisms engaged by men and women when solving the same decision-making task are different. These observations indicate that sex-related differences contribute to the heterogeneity observed in both normal and abnormal brain functioning. These results also provide further evidence of sexual dimorphism in neurocognitive performance and brain function.

## Introduction

The existence of sex-related differences in decision-making and the identification of their neurological correlates will have significant impact on behavioral and neuroimaging studies in the setting of clinical research. Although the majority of earlier investigations have involved only men, recent governmental and ethical mandates have led to a greater inclusion of women in clinical research (General Accounting Office, 1992). This has led to a greater emphasis on discovering potential differences in sex-dependent etiological factors of diseases (Waldron, 1983), aging and longevity (Murphy et al., 1996). Neurobehavioral studies show that men outperform women on tasks of reaction time (Bleecker et al., 1987), spatial reasoning (Resnick, 1993) and decision-making (Reavis and Overman, 2001). In contrast, women outperform men on tasks of verbal memory (Bolla-Wilson and Bleecker, 1986; Ragland et al., 2000), eye–hand coordination (Agnew et al., 1988) and verbal fluency (Bolla et al., 1998). These sex-related differences may be related to differences in brain function and structure (Rodriguez et al., 1988; Andreason et al., 1994; Azari et al., 1995; Esposito et al., 1996; Murphy et al., 1996). Although neuroimaging studies are very sensitive in detecting alterations in brain activity, findings from these studies are also impacted by a high degree of variability stemming from innate individual differences. Since most neuroimaging studies combine men and women in unequal ratios into groups, it is critical to determine if brain imaging and behavioral results can be generalized from men to women and vice versa. Indeed, a limited number of studies with neuroimaging have reported important sex-related differences in cognition and brain functioning (Azari et al., 1992; Esposito et al., 1996; Kawachi et al., 2002). Thus, in men, higher relative metabolic activity in the right hemisphere (RH) was associated with better performance on cognitive tasks reflecting RH function, and higher relative left hemispheric (LH) metabolic activity was associated with better performance on LH-mediated cognitive tasks (Azari et al., 1995). In contrast, in women, higher metabolic rates in a specific hemisphere were associated with better performance on tasks reflecting contralateral hemisphere functioning (Azari et al., 1995). These results suggest greater hemispheric functional specialization in men and greater interhemispheric transfer in women (Azari et al., 1995).

Using similar neuroimaging technology, we investigated sex-related differences in brain activity in the orbitofrontal cortex (OFC) during performance of a decision-making task. This interest stemmed from several sources. First, in non-human primates, the OFC develops more rapidly in male infants and males perform better than female infant monkeys on an object reversal task that utilizes the OFC (Goldman et al., 1974; Clark and Goldman-Rakic, 1989). Second, the OFC together with the dorsolateral prefrontal cortex (DLPFC) comprise part of a neural network that mediates performance on the Iowa Gambling Task (Bechara et al., 1994, 2000a,b; Ernst et al., 2002; Manes et al., 2002; Bolla et al., 2003). The Iowa Gambling Task is frequently used to assess decision-making and OFC functioning in patients with lesions to the ventromedial prefrontal cortex (VMPFC) (Bechara et al., 1994) and substance abusers (Bolla et al., 2003). Third, men out-performed women on the Iowa Gambling task (Reavis and Overman, 2001).

This study was designed to investigate sex-related differences in performance and brain activity in the OFC during performance of a decision making task. On the basis of previous studies (Reavis and Overman, 2001) we predicted (i) that men would perform better than women on the computerized version of the Iowa Gambling Task; and (ii) that, based on our previous work (Ernst et al., 2002; Bolla et al., 2003), men would show patterns of activation different from those of women in the OFC and DLPFC during performance of this task.

## Materials and Methods

### Research Participants

Twenty healthy subjects participated in this study (10 women and 10 men). All participants received a full medical and psychiatric screening with the Psychiatric Diagnostic Interview Schedule (DIS) (Robins et al., 1981). Inclusion criteria were age between 21 and 45, IQ > 80 (assessed by the Shipley Institute of Living Scale) (Zachary, 1991), right handedness, English as a first language, no past history or current use of illicit drugs (screened by urine toxicology), as assessed by the Drug Use Survey Questionnaire (Smith, 1991) and the Addiction Severity Index (McLellan et al., 1980), and consumed <10 alcoholic drinks/week. Participants were excluded if they reported past or current Axis I disorder other than nicotine dependence by DSM-IV criteria according to the DIS (e.g. anxiety disorder or major depressive disorder), evidence of acute or chronic medical problems, or positive pregnancy test for women. This study was approved by the institutional review boards at the Johns Hopkins Medical Institutions Joint Committee on Clinical Investigation and the Johns Hopkins Bayview Medical Center. The study was also approved by National Institute on Drug Abuse — Intramural Research Program (NIDA-IRP). All volunteers gave written informed consent and were compensated for their time. Volunteers were recruited through newspaper advertisements.

### Data Collection

After successfully completing the consent process, participants were admitted to our General Clinical Research Center (GCRC) for a 3 day inpatient stay. Over this period they also participated in other aspects of a larger study that will be reported elsewhere. This inpatient stay ensured that participants received adequate nutrition and rest.

### Design

For the positron emission tomography (PET) session (on day 3 of the residential stay) participants received six injections of H215O water (10 mCi each). A 1 min acquisition scan was collected after the injection. Three cognitive conditions were studied: rest (R; eyes fixated on a target); active task (A; Iowa Gambling Task); and control task (C; sensorimotor control task). Two scans were acquired for each of the three conditions. Task order was counterbalanced within and between participants. The tasks began 1 min before injection of the tracer to ensure that the participant was cognitively engaged in the task at the time of image acquisition. The task ended when the participant had selected 100 cards (∼6 min). Before the actual scanning, participants performed a practice task to acclimate them to responding with the mouse while in the scanner. A molded facemask was created to minimize head movement during scan acquisition, limiting the participant’s ability to speak without hindering vision or manual responding. Stimuli were presented at the center of an LCD monitor, controlled by a Toshiba laptop computer. The monitor was mounted ∼1 m above and in front of the participant, tilted ∼40° from the bed so that the participant could view the screen from inside the scanner when wearing the facemask. Participants were instructed to abstain from smoking cigarettes and caffeinated beverages for 3 h before the study.

Since decision-making and OFC function are frequently studied in neuropsychiatric disorders such as substance abuse (Bolla et al., 2003), the Iowa Gambling Task was used as our activation task. Lesion and neuroimaging studies indicate that the right OFC is a primary component in performing well on this task (Bechara et al., 2000a,b; Ernst et al., 2002; Bolla et al., 2003). The task evaluates decision-making by measuring the participant’s ability to choose between high gains with a risk of extremely high losses (negative net score), and low gains with a risk of smaller losses (positive net score). Participants were instructed to win as much money as possible by picking one card at a time from each of four decks (A, B, C and D) in any order until the computer instructed them to stop (after the selection of the 100th card). They were also told that some decks are worse than others and that they would win if they stayed away from the worse decks. While performing the task participants were informed of the amount of money they had remaining after each card was selected. Participants selected on average ∼17 cards/min. A net global outcome score (net score) was calculated by subtracting the total number of cards selected from the disadvantage decks (A + B) from the total number of cards selected from the advantage decks (C + D) in trials 1 and 2, and then deriving a mean for both trials. To motivate participants to perform well on this task they were informed that for each ‘game dollar’ they won, we would pay them one cent ‘real money’ equaling a possible $20.00 per trial and maximum total of$40.00 for both trials.

The control task was designed to be analogous to the Iowa Gambling Task with respect to sensorimotor demands and exposure to gains and losses. Unlike the active task, the gains and losses associated with the control task were equal between decks and participants were instructed to select cards sequentially in the fixed order of A-B-C-D-A-B-C-D, etc. Consequently, card selection did not require decision-making. Although the sensory and motor aspects of the control task were identical to those in the active task, rewards and penalties were contingent upon behavior in the active task but non-contingent in the control task.

### PET Scan Acquisition

Scans were acquired with a Siemens ECAT EXACT HR +, in 63 planes with a 15.5 cm field of view in 3-D mode. Images were reconstructed using a Hann filter with 0.5 cut off frequency. The average transverse resolutions (full width half maximum (FWHM)) of the scanner at 1 and 10 cm from the center of the field of view, measured in 3-D mode and determined using a fluorine-18 line source and a ramp filter (with a 0.5 cutoff frequency), were 4.66 and 5.45 mm, respectively. Axial resolutions of the scanner (FWHM), measured using a point source of 18F and the same reconstruction algorithm were 4.21 and 5.0 mm at 0 and 10 cm from the center, respectively. In case of application of a Hann filter with a 0.5 cutoff frequency, used for reconstruction of brain images, the average transverse resolutions were 6.52 and 7.16 mm, respectively. For the same reconstruction algorithm, the average axial resolution at 0 cm from the center was 3.72 mm and at 10 cm, 5.64 mm.

### Image Processing and Statistical Analyses

PET images were realigned, spatially normalized into the Montreal Neurological Institute coordinate system, and smoothed with a 12 × 12 × 12 mm Gaussian kernel by using Statistical Parametric Mapping Software (SPM 99; Welcome Department of Cognitive Neurology). A two-stage procedure was implemented for statistical analyses for within-group effects (n = 10) and between-group effects (n = 20). In the first stage, PET images from each participant were used to create an individual adjusted mean image, representing the relative change in brain activity (normalized rCBF) between the active and control tasks (all scans from the active task minus the control task). Thus, the adjusted mean image represents the change in brain activation between the active task and the control task. This change in brain activity was taken to reflect the process of decision-making by subtracting the motor, auditory and visual components involved in the task (control task) from the higher cognitive functions of decision-making (active task). Proportional scaling was used to correct for within-session variations in global signal for each adjusted mean image. Importantly, no significant differences were found in the correlation between global signal and the conditions of interest (Worsley et al., 1996; Andersson, 1997; Desjardins et al., 2001). To examine within-subject effects, we then entered the adjusted image from each participant into a random effects one-sample t-test (n = 10) with 9 degrees of freedom (df). Stage two of the procedure to examine between-group differences involved entering the adjusted mean image for each participant into a random effects two-sample t-test (n = 20) with 18 df.

To examine our specific regions of interest (OFC and DLPFC) and correct for multiple comparisons, we employed the ‘small volume correction’ method featured in SPM99 using our own voxel volume of interest (VVOI) image templates (Matochik et al., 2003). This method did not restrict the search volume for statistical analyses to a single stereotaxic x–y–z coordinate point taken from previous findings in the literature, but rather used the actual spatially normalized volume of the region of interest to limit the search. For example, the entire lateral OFC region volume is searched rather than an area within a sphere or box centered around single coordinate that may be on or near the region of interest. This procedure ensured that our experiment-wise false positive rate (Type I error) for a particular region of interest was maintained at the α < 0.05 level. Bilateral small volume templates were constructed for the following brain regions: orbitofrontal cortex [medial region (Brodmann area BA 11), including the gyrus rectus and medial orbital gyrus; and the lateral orbital region (BA 11, 47), including lateral orbital gyrus and a portion of the inferior frontal gyrus], and the lateral prefrontal cortex [middle and inferior frontal gyrus (BA 9, 10, 44, 45, 46)].

## Results

### Demographics

The group of men (n = 10) was matched to the group of women (n = 10) on the Shipley IQ score. There were no significant group differences in age, years of education, maternal education, Shipley IQ, Hollingshead Index of Socioeconomic Status, race, alcohol use or proportion of smokers (Table 1).

Figure 1 shows the results of the Mann–Whitney U-test that was used to test for group differences between men and women in net outcome score (cards from advantage decks minus cards from disadvantage decks; the higher the score the better the performance; mean ± SD net score: men, 25.2 ± 14.8; women, –12.2 ± 25.3). Men performed significantly better than women on the task (Z = –2.86; P < 0.01). We also examined sex-related differences in performance between the first and second trials (learning) of the task using a Wilcoxon Signed Rank test. Men significantly increased performance from trial 1 to trial 2 (trial 1: mean = 8.2 ± 17.7, trial 2: mean = 42.2 ± 22.0, Z = –2.55, P < 0.01; women did not show this significant increase (trial 1: mean = –15.0 ± 22.1, trial 2: mean = –9.4 ± 33.0, Z = –0.71, P = 0.47).

Figure 2 shows individual performance results for men and women separately. Only one of nine men received a negative score on the task indicating selection of more cards from the disadvantage decks than the advantage decks. In contrast, five of ten women received negative scores and another three yielded scores of ≤5. Since the Iowa Gambling Test is self-paced, we also examined differences in rate of responding between men and women. No significant differences were found in the rate of responding. The rate of responding (mean time to complete the 100 card task) was: 5:80 ± 1:16 min for women and 6:98 ± 2:21 min for men [t(17) = –1.48, P = 0.16].

### Brain Activations in Women and Men during Performance of the Iowa Gambling Task

Patterns of activation during task performance were examined for men and women in a random effects one-sample t-test. When voxel volume of interest (VVOI) templates were applied to correct for multiple comparisons, men showed significant activation in two large clusters, one in the right lateral OFC (845 contiguous voxels) and one in the left lateral OFC (217 contiguous voxels) whereas women showed a single smaller significant cluster in the left medial OFC (173 voxels activated) (Table 2, Fig. 3). However, when the entire brain was searched for areas of activation, men showed large areas of activation in an additional region of the right lateral OFC, two regions in the right DLPFC, and the right parietal lobe (Table 2).

A laterality index [(number of voxels significantly activated; right hemisphere – left hemisphere)/(total voxels) × 100] was computed for women and men separately (Christodoulou et al., 2001). (A positive score indicates greater RH activation, whereas a negative score indicates greater LH activation.) Men demonstrated extreme RH lateralization with a score of 73, whereas women did not demonstrate either RH or LH lateralization with a score of only 6.

### Differences between Men and Women in Brain Activity during Performance on the Iowa Gambling Task

When VVOI templates were applied to correct for multiple comparisons, men showed significantly greater activation than women in the right lateral OFC. Women showed greater activation than men in the left DLPFC (Table 3, Fig. 4). Post hoc between-group exploratory analyses of the entire brain determined that women had greater activation than men in the left medial frontal gyrus and left temporal lobe [t(18) ≥ 3.61, height threshold P < 0.001, extent 50 contiguous voxels, uncorrected; Table 3].

We determined if random feedback in the control task engaged the OFC more in women than in men, resulting in a ceiling effect. If this were the case, women could not show additional activation during the gambling task. We compared men and women on the both the resting condition and the control condition and found no significant differences in right OFC activation in either the resting and control conditions.

## Discussion

As expected, men performed significantly better than women on the Iowa Gambling Task. This finding replicates previous reports of sex-related performance differences (men better than women) on this task (Reavis and Overman, 2001). Our neuroimaging observations also provide further insight into the mechanistic bases for the better performance by men on this specific decision-making task. The within-group analyses showed that men activated large regions of the lateral OFC bilaterally, RH more than LH, right DLPFC and right parietal lobe, whereas women only activated a smaller region in the left medial OFC during task performance. This appears to be very advantageous for men because the Iowa Gambling Task relies more heavily on the integrity of the right OFC than the left OFC (Manes et al., 2002; Tranel et al., 2002). On the other hand, women activated the left DLPFC significantly more than men. These issues are discussed in more detail below.

In addition to better mean performance, men also showed a significant learning effect from trial 1 to trial 2, whereas women failed to show similar improvements (Fig. 1). These results suggest that women may be using less effective strategies, given their initial poor performance and their failure to improve during a second trial. Thus, brain mechanisms engaged by men and women when solving the same decision-making task are different, which appears to put women at a relative disadvantage. This view is supported by epidemiological data that show pathological gambling to be less prevalent in women. Because only 30% of pathological gamblers are women (Coventry and Constable, 1999), it is not far-fetched to suggest women gamblers might perform similarly to men on this task. This remains to be tested.

Since the pattern of performance and brain activity was so different between men and women we were concerned that our group of women was in some way unusual. To examine this we compared the performance and brain activity of the exact same women and men used in the present study on another task, the Stroop. We found no sex-related differences in either performance or brain activity (also assessed by PET), suggesting these results are specific to the Iowa Gambling Task and not an artifact of participant selection. Also, in a study using the same version of the task that we used, mean group scores of patients with different brain lesions ranged from 23.6 ± 25 to –9.5 ± 18 (Clark et al., 2003). Our mean values are comparable to their values (mean net scores: men = 25.2 ± 14.8, women = –12.2 ± 25.3). Although age was not significantly different between men and women there was however a five year age difference between the groups. Nevertheless, we do not believe that this difference in age contributed to the large performance difference between groups since age was not correlated with task performance (r = 0.17). Nevertheless, these results should be replicated in a larger sample.

Brain activity in men was lateralized almost exclusively to the RH and men showed significantly more activation in the right lateral OFC than women. Based on a variety of sources, we believe that these findings are biologically plausible since they replicate and extend the work of others. First, findings from ventral medial prefrontal cortex lesion studies reveal that impairment on this task and with decision-making in real-life primarily occurs after right OFC unilateral lesions, while left unilateral lesions appear to have little effect on decision-making (Bechara et al., 2000a,b; Manes et al., 2002; Tranel et al., 2002). The lateral OFC is sensitive to punishment and overrides behavior based on the previous rewarding values of stimuli and responses, whereas the medial OFC is involved in reward and guessing situations when the outcomes are undetermined (O’Doherty et al., 2001). It is probable that better task performance in men may be directly related to more substantial activation in the right lateral OFC, a sub-region that appears to be the crucial in punishment and the higher cognitive demands of decision-making. Therefore, increased lateral OFC activation in men could be a representation of the punishing consequences of selecting cards from the bad deck, and thus a substrate of good decision-making, or alternatively could simply be a response to more positive feedback. We speculate that the first interpretation is more accurate since other studies showed increased activation in the lateral OFC (bilaterally) in response to a punishing outcome (O’Doherty et al., 2001). In addition, the OFC has been shown to mediate the valence of olfactory stimuli. For example, the right OFC was shown to be associated with pleasant olfactory stimulation while activation in the left OFC was associated with unpleasant olfactory stimulation (Anderson et al., 2003). It is therefore likely that men and women perceive the valence of the Iowa Gambling Task differently, which would result in different patterns of brain activity. In order to identify the aspect(s) of decision-making that is associated with greater OFC activation, experimental conditions within a decision-making task will need to be manipulated in future studies. Second, other investigators have reported that men have higher activity in the RH, activity that is associated with better performance on ‘RH’ cognitive tasks (Azari et al., 1995). In contrast, women show higher brain activity in a specific hemisphere that is associated with better cognitive performance on tests reflecting functioning of the contralateral hemisphere (Azari et al., 1995). In our study, women showed greater activation in the LH (OFC and DLPFC) while performing a task that is predominantly an RH task (Ernst et al., 2002). Therefore, as suggested by others (Azari et al., 1995), men may have greater hemisphere functional specialization and women may have greater inter-hemispheric transfer. Third, men show greater right than left frontal asymmetry that is not found in women (Rodriguez et al., 1988). Finally, men have more lateralized brain activation than women, and women have more diffuse brain activation than men (Kawachi et al., 2002; Rodriguez et al., 1988; Rossell et al., 2002). All of these observations were confirmed in our current study.

These results have important clinical implications and provide evidence to support sexual dimorphism of the brain. Our results show that women are not as adept as men in decision-making when performing the Iowa Gambling Task. In addition, these sex-related differences in performance are related to differential brain functioning. These observations indicate that women utilize different cognitive strategies and alternative neural networks when performing this specific task. These findings illustrate the usefulness of combining neurocognitive measures with functional neuroimaging techniques to investigate sex differences in cognitive processing and to understand the mechanisms behind them.

Supported by NIH grants DA 11426 (K.B.) and the JHBMC-GCRC (MO1 RR02719) and the DHHS/NIH/NIDA Intramural Research Program. We would like to thank all the nurses and staff at NIDA-IRP, the Brain Imaging Center and the Bayview GCRC who contributed to this project. We especially thank Kent A. Kiehl, Ph.D. for data analyses support, Debra Hill, B.A. for computer and database support, and Pamela Talalay, Ph.D. for editorial guidance.

Address correspondence to Karen I. Bolla, Ph.D., Johns Hopkins Bayview Medical Center, Department of Neurology, 4940 Eastern Avenue, Baltimore, MD 21224, USA. Email: kbolla@jhmi.edu.

Figure 1. Performance differences between men and women on the Iowa Gambling Task, based on the mean performance scores (net score = number of cards from the advantage decks minus number of cards from the disadvantage decks) of men and women on the task (left figure). The higher the score the better the performance. Learning effects (performance of trials 1 and 2) in men and women (right figure); **P < 0.01.

Figure 1. Performance differences between men and women on the Iowa Gambling Task, based on the mean performance scores (net score = number of cards from the advantage decks minus number of cards from the disadvantage decks) of men and women on the task (left figure). The higher the score the better the performance. Learning effects (performance of trials 1 and 2) in men and women (right figure); **P < 0.01.

Figure 2. The Iowa gambling task performance in men and women, presented as mean performance scores for men (n = 9) and women (n = 10). Data points overlap at a mean score of –26 for the women and for the men (n = 9) because of missing data for one participant. Negative performance indicates that more cards were chosen from the disadvantage decks than the advantage decks.

Figure 2. The Iowa gambling task performance in men and women, presented as mean performance scores for men (n = 9) and women (n = 10). Data points overlap at a mean score of –26 for the women and for the men (n = 9) because of missing data for one participant. Negative performance indicates that more cards were chosen from the disadvantage decks than the advantage decks.

Figure 3. Brain activity in men and women during performance on the Iowa Gambling Task, shown as statistical parametric maps showing (glass brains) brain activation during performance of the Task (active task minus control task). Maps of the t-statistics showing all voxels significantly activated at P < 0.001 (peak) within a cluster extent threshold of k > 40. All images are in neurological convention (right is right). (a) Men, regional activation includes the right lateral OFC (BA 47), two regions of the right DLPFC (BA 9, 10) and the right parietal lobe (BA 40). (b) Women, regional activation includes the left medial OFC (BA 11).

Figure 3. Brain activity in men and women during performance on the Iowa Gambling Task, shown as statistical parametric maps showing (glass brains) brain activation during performance of the Task (active task minus control task). Maps of the t-statistics showing all voxels significantly activated at P < 0.001 (peak) within a cluster extent threshold of k > 40. All images are in neurological convention (right is right). (a) Men, regional activation includes the right lateral OFC (BA 47), two regions of the right DLPFC (BA 9, 10) and the right parietal lobe (BA 40). (b) Women, regional activation includes the left medial OFC (BA 11).

Figure 4. Group differences in brain activation in men and women during performance on the Iowa Gambling Task. All images are in neurological convention (right is right). (a) Men showed more activation than women in the right OFC [peak located at 42, 41, –4 (BA 47, 10); P < 0.05, k = 246 voxels, extent corrected] during task performance (active task minus control task). (b) Women showed greater activation than men in the left DLPFC [peak located at –26, 29, 34 (BA 9); P < 0.01, k = 20, height uncorrected].

Figure 4. Group differences in brain activation in men and women during performance on the Iowa Gambling Task. All images are in neurological convention (right is right). (a) Men showed more activation than women in the right OFC [peak located at 42, 41, –4 (BA 47, 10); P < 0.05, k = 246 voxels, extent corrected] during task performance (active task minus control task). (b) Women showed greater activation than men in the left DLPFC [peak located at –26, 29, 34 (BA 9); P < 0.01, k = 20, height uncorrected].

Table 1

Demographic characteristics for both men and women

 Characteristic Men (n = 10) Women (n = 10) Age (years) 32.6 ± 5.9 (24–42) 27.5 ± 6.3 (21–42) Education (years) 13.9 ± 3.0 (10–19) 13.9 ± 2.6 (8–18) Mother’s years of education 13.5 ± 2.8 (9–18) 14.0 ± 2.4 (9–16) Shipley IQ 97.1 ± 8.1 (83–110) 101.7 ± 8.70 (89–112) Hollingshead SES 3.9 ± 1.3 (1–5) 4.3 ± 1.1 (2–5) Race (A.A./C.) 6/4 0 Hispanic 5/4 1 Hispanic Alcohol use (drinks/week) 1.2 ± 1.9 (0–6) 1.8 ± 2.1 (0–6) Cigarette Smokers 7/10 5/10
 Characteristic Men (n = 10) Women (n = 10) Age (years) 32.6 ± 5.9 (24–42) 27.5 ± 6.3 (21–42) Education (years) 13.9 ± 3.0 (10–19) 13.9 ± 2.6 (8–18) Mother’s years of education 13.5 ± 2.8 (9–18) 14.0 ± 2.4 (9–16) Shipley IQ 97.1 ± 8.1 (83–110) 101.7 ± 8.70 (89–112) Hollingshead SES 3.9 ± 1.3 (1–5) 4.3 ± 1.1 (2–5) Race (A.A./C.) 6/4 0 Hispanic 5/4 1 Hispanic Alcohol use (drinks/week) 1.2 ± 1.9 (0–6) 1.8 ± 2.1 (0–6) Cigarette Smokers 7/10 5/10

Numbers in parentheses are ranges.

Table 2

Brain activations in men and women separately during performance on the Iowa Gambling Task

 Side Talairach coordinates BA Voxel T Cluster size k x y z Men A priori regions Lateral OFC R 38 44 –4 47 6.29* 845 Lateral OFC L –34 54 –16 11 7.10* 217 Entire brain w/o (ROI–SVC) Lateral OFC R 18 38 –20 47 6.32** 285 DLPFC R 30 47 1 10 8.81** 242 DLPFC R 16 40 22 9 7.11** 242 Parietal lobe R 50 –42 52 40 6.68** 181 Women A priori regions Medial OFC L –12 40 –14 11 4.38** 173
 Side Talairach coordinates BA Voxel T Cluster size k x y z Men A priori regions Lateral OFC R 38 44 –4 47 6.29* 845 Lateral OFC L –34 54 –16 11 7.10* 217 Entire brain w/o (ROI–SVC) Lateral OFC R 18 38 –20 47 6.32** 285 DLPFC R 30 47 1 10 8.81** 242 DLPFC R 16 40 22 9 7.11** 242 Parietal lobe R 50 –42 52 40 6.68** 181 Women A priori regions Medial OFC L –12 40 –14 11 4.38** 173

For a priori regions using VVOI templates for small volume correction: random effects, peak height threshold P < 0.01, extent threshold = 0, VVOI templates corrected for multiple comparisons bilaterally within the lateral OFC, medial OFC, and DLPFC, with 9 df. *P < 0.05, corrected for height threshold.

For exploratory analysis of entire brain (without small volume correction): **P < 0.01, corrected for extent threshold.

Table 3

Group differences in brain activity during performance on the Iowa Gambling Task

 Side Talairach Coordinates BA Voxel T Cluster size k x y z Men > women A priori region Lateral OFC R 42 41 –4 47, 10 4.05* 246 Women > men A priori region DLPFC L –26 29 34 9 3.19** 20 Entire brain Medial frontal gyrus L –12 –1 52 6 4.63** 74 Temporal lobe L –42 –7 –20 20 4.80** 57
 Side Talairach Coordinates BA Voxel T Cluster size k x y z Men > women A priori region Lateral OFC R 42 41 –4 47, 10 4.05* 246 Women > men A priori region DLPFC L –26 29 34 9 3.19** 20 Entire brain Medial frontal gyrus L –12 –1 52 6 4.63** 74 Temporal lobe L –42 –7 –20 20 4.80** 57

For a priori regions using VVOI templates for small volume correction: random effects, peak height threshold P < 0.01, extent threshold = 0. VVOI templates corrected for multiple comparisons bilaterally within the lateral OFC, medial OFC, and DLPFC, with 18 df.

For exploratory analysis of entire brain, no small volume correction was applied. *P < 0.05, corrected at the height threshold level; **P < 0.005, uncorrected at the height threshold level.

## References

Agnew J, Bolla-Wilson K, Kawas C, Bleecker ML (
1988
) Purdue Pegboard age and sex norms for ages forty and older.
Dev Neuropsychol

4
:
29
–35.
Anderson AK, Christoff K, Stappen I, Panitz D, Ghahremani DG, Glover G, Gabrieli JDE, Sobel, N (
2003
) Dissociated neural representations of intensity and valence in human olfaction.
Nat Neurosci

6
:
196
–202.
1997
) How to estimate global activity independent of changes in local activity.
Neuroimage

6
:
237
–244.
Andreason PJ, Zametkin AJ, Guo AC, Baldwin P, Cohen RM (
1994
) Gender-related sex differences in regional cerebral glucose metabolism in normal volunteers.
Psychiatry Res

51
:
175
–183.
Azari NP, Rapoport SI, Grady CL, DeCarli C, Haxby JV, Schapiro MB, Horwitz B (
1992
) Gender differences in correlations of cerebral glucose metabolic rates in young normal adults.
Brain Res

574
:
198
–208.
Azari NP, Pettigrew KD, Pietrini P, Murphy DG, Horwitz B, Schapiro MB (
1995
) Sex differences in patterns of hemispheric cerebral metabolism: a multiple regression/discriminant analysis of positron emission tomographic data.
Int J Neurosci

81
:
1
–20.
Bechara A, Damasio AR, Damasio H, Anderson SW (
1994
) Insensitivity to future consequences following damage to human prefrontal cortex.
Cognition

50
:
7
–15.
Bechara A, Tranel D, Damasio H (
2000
) Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions.
Brain

123
:
2189
–2202.
Bechara A, Damasio H, Damasio AR (
2000
) Emotion, decision making and the orbitofrontal cortex.
Cereb Cortex

10
:
295
–307.
Bleecker ML, Bolla-Wilson K, Agnew J, Meyers DA (
1987
) Simple visual reaction time: sex and age differences.
Dev Neuropsychol

3
:
165
–172.
Bolla-Wilson K, Bleecker ML (
1986
) Influence of verbal intelligence, sex, age, and education on the Rey Auditory Verbal Learning Test.
Dev Neuropsychol

2
:
203
–211.
Bolla KI, Gray S, Galante R, Resnick SM, Kawas C (
1998
) Category and letter fluency in highly educated older adults.
Clin Neuropsychol

12
:
330
–338.
Bolla KI, Eldreth DA, London ED, Kiehl KA, Mouratidis M, Contoreggi C, Matochik JA, Kurian V, Cadet JL, Kimes AS, Funderburk FR, Ernst M (
2003
) Orbitofrontal cortex dysfunction in abstinent cocaine abusers performing a decision-making task.
Neuroimage

19
:
1085
–1094.
Christodoulou C, DeLuca J, Ricker JH, Madigan NK, Bly BM, Lange G, Kalnin AJ, Liu WC, Steffener J, Diamond BJ, Ni AC (
2001
) Functional magnetic resonance imaging of working memory impairment after traumatic brain injury.
J Neurol Neurosurg Psychiatry

71
:
161
–168.
Clark A, Goldman-Rakic PS (
1989
) Gonadal hormones influence the emergence of cortical function in nonhuman primates.
Behav Neurosci

103
:
1287
–1295.
Clark A, Manes F, Antoun N, Sahakian BJ, Robbins TW (
2003
) The contributions of lesion lterality and lesion volume to decision-making impairment following frontal lobe damage.
Neuropsychologia

41
:
1474
–1483.
Coventry KR, Constable B (
1999
) Physiological arousal and sensation-seeking in female fruit machine gamblers.

94
:
425
–432.
Desjardins AE, Kiehl KA, Liddle PF (
2001
) Removal of confounding effects of global signal in functional magnetic resonance imaging analyses.
Neuroimage

13
:
751
–758.
Ernst M, Bolla K, Mouratidis M, Contoreggi C, Matochik JA, Kurian V, Cadet JL, Kimes AS, London ED (
2002
) Decision-making in a risk-taking task: a PET study.
Neuropsychopharmacology

26
:
682
–691.
Esposito G, Van Horn JD, Weinberger DR, Berman KF (
1996
) Gender differences in cerebral blood flow as a function of cognitive state with PET.
J Nucl Med

37
:
559
–564.
General Accounting Office (
1992
) Women’s health: FDA needs to ensure more study of gender differences in prescription drug testing. GAO/HRD-93-17. Washington, DC: Government Printing Office.
Goldman PS, Crawford HT, Stokes LP, Galkin TW, Rosvold H (
1974
) Sex-dependent behavioral effects of cerebral cortical lesions in the developing rhesus monkey.
Science

186
:
540
–542.
Grant S, Contoreggi C, London ED (
2000
) Drug abusers show impaired performance in a laboratory test of decision-making.
Neuropsychologia

38
:
1180
–1187.
Kawachi T, Ishii K, Sakamoto S, Matsui M, Mori T, Sasaki M (
2002
) Gender differences in cerebral glucose metabolism: a PET study.
J Neurol Sci

199
:
79
–83.
MacDonald AW, Cohen JD, Stenger VA, Carter CS (
2000
) Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control.
Science

288
:
1835
–1838.
Manes F, Sahakian B, Clark L, Rogers R, Antoun N, Aitken M, Robbins T (
2002
) Decision-making processes following damage to the prefrontal cortex.
Brain

125
:
624
–639.
Matochik JA, London ED, Eldreth D, Cadet J-L, Bolla K (
2003
) Frontal cortical tissue composition in abstinent cocaine abusers: a magnetic resonance imaging study.
Neuroimage

19
:
1095
–1102.
McLellan AT, Luborsky L, Woody GE, O’Brien C (
1980
) An improved evaluation instrument for substance abuse patients: the Addiction Severity Index.
J Nervous Mental Disord

168
:
26
–33.
Murphy DG, DeCarli C, McIntosh AR, Daly E, Mentis MJ, Pietrini P, Szczepanik J, Schapiro MB, Grady CL, Horwitz B, Rapoport SI (
1996
) Sex differences in human brain morphometry and metabolism: an in vivo quantitative magnetic resonance imaging and positron emission tomography study on the effect of aging.
Arch Gen Psychiatry

53
:
585
–594.
O’Doherty J, Kringelbach ML, Rolls ET, Hornak J, Andrews C (
2001
) Abstract reward and punishment representations in the human orbitofrontal cortex.
Nat Neurosci

4
:
95
–102.
Ragland JD, Coleman AR, Gur RC, Glahn DC, Gur RE (
2000
) Sex differences in brain–behavior relationships between verbal episodic memory and resting regional cerebral blood flow.
Neuropsychologia

38
:
451
–461.
Reavis R, Overman WH (
2001
) Adult sex differences on a decision-making task previously shown to depend on the orbital prefrontal cortex.
Behav Neurosci

115
:
196
–206.
Resnick SM (
1993
) Sex differences in mental rotations: an effect on time limits?
Brain Cogn

21
:
71
–79.
Robins LN, Helzer JE, Croughan J(
1981
) National Institute of Mental Health Diagnostic Interview Schedule: its history, characteristics, and validity.
Arch Gen Psychiatry

38
:
381
–389.
Rodriguez G, Warkentin S, Risberg J, Rosandi G (
1988
) Sex differences in regional cerebral blood flow.
J Cereb Blood Flow Metab

8
:
783
–789.
Rossell SL, Bullmore ET, Williams SCR, David AS (
2002
) Sex differences in functional brain activation during a lexical visual field task.
Brain Lang

80
:
97
–105.
Smith SS (
1991
Tranel D, Bechara A, Denburg NL (
2002
) Asymmetric functional roles of right and left ventromedial prefrontal cortices in social conduct, decision-making, and emotional processing.
Cortex

38
:
589
–612.
Waldron I (
1983
) Sex differences in illness incidence, prognosis, and mortality.
Soc Sci Med

17
:
1107
–1123.
Worsley KJ, Marrett S, Neelin P, Vandal AC, Friston KJ, Evans AC (
1996
) A unified statistical approach for determining significant signals in images of cerebral activation.
Hum Brain Mapp

4
:
58
–73.
Zachary RA (
1991
) Shipley Institute of Living Scale: revised manual. Los Angeles, CA: Western Psychological Services.