In the 1970s, Walter Mischel created the delay of gratification paradigm to test children's ability to resist temptation and forgo a small immediate reward in order to obtain a larger delayed reward. The paradigm consists of preschool children being asked to observe one marshmallow, without touching or eating it. The research assistant leaves the child alone and if the child can resist the temptation to touch the marshmallow and is able to wait until the research assistant returns voluntarily, the child is given the opportunity to have two marshmallows. This elegant and simple paradigm allows researchers to test children's ability to select a large delayed reward in favor of small immediate reward (Mischel and Underwood, 1974). Presumably, children who are capable of waiting and resisting the temptation of the small immediate reward have higher self-control and are better able to delay gratification.

According to the hypothesis proposed by Bjorklund and Kipp (1996), one might expect that females will be better able to delay gratification and resist a small immediate reward in comparison to males. Interestingly, Mischel and Underwood (1974) were the first to report a sex difference in delay of gratification. The authors reported that female preschoolers were able to wait for significantly longer periods of time to obtain the larger reward in comparison to their male counterparts—these results confirm the predictions proposed by Bjorklund and Kipp (Mischel and Underwood, 1974). More recently, a meta-analysis was conducted on experiments using the delay of gratification paradigm with the intention of directly testing Bjorklund and Kipp's (1996) hypothesis that there is a female advantage in inhibitory control. Silverman (2003) investigated 33 studies that required participants to make a choice between a small immediate reward and larger delayed reward. The results of the meta-analysis revealed a statistically significant female advantage in the capacity to delay gratification. Silverman (2003) concludes that the effect size of the female advantage in delay of gratification is relatively small; however, this may be due to instruments that lack precision and small sample sizes. Importantly, the author states that these results provide substantial support for the hypothesis proposed by Bjorklund and Kipp (Silverman, 2003). Based on these findings, it appears as though females are better suited to delay gratification and resist temptation in comparison to males and that this sex difference emerges early in development.

Sex differences in the structure and function of the MCLP might provide some insight into the inconsistent findings on self-regulation and inhibition at the behavioral level. Much of the research investigating sex differences in the MCLP focuses on reward-related activity primarily because the MCLP is involved in rewarding and motivated behaviors. It is conceivable that individuals who have increased neural responses to rewards and/or are more sensitive to the prospect of obtaining rewards would have greater difficulty inhibiting prepotent responses and delaying gratification.

The MCLP originates in the ventral tegmental area (VTA) and has projections to the nucleus accumbens (NAc), bed nucleus of stria terminalis (BNST), and the frontal cortex. GABAergic, glutamatergic, and opioid peptidergic systems modulate activity in the VTA and NAc (Bobzean et al., 2014). Furthermore, dopaminergic transmission within the MCLP modulates rewarding and motivated behaviors. Sexual dimorphisms within the structure and function of the MCLP have been well documented in both human and animal samples (See Table 3). For example, a recent Magnetic Resonance Imaging (MRI) study that investigated sexual dimorphisms in the basal ganglia revealed that sexual dimorphisms exist in some structures of the basal ganglia (globus pallidus and putamen) but not others (caudate nucleus and nucleus accumbens) (Rijpkema et al., 2012). Andersen et al. (1997) also investigated structural sex differences in the MCLP using audioradiography and demonstrated developmental sexual dimorphisms in D1 and D2 receptors the striatum of male and female rats. Activation of D1 receptors ultimately increased cyclic adenosine monophosphate (cAMP), whereas activation of D2 receptors indirectly reduced cAMP concentrations. Specifically, males had enhanced overproduction and elimination of D1 and D2 receptors relative to females, and males sustained the greater D1 overproduction in the nucleus accumbens into adulthood (Andersen et al., 1997). Evidently, structural sexual dimorphisms exist in the development of the MCLP.

In addition to the above purely structural studies, numerous other research groups have investigated both structural and functional sex differences. For example, Pohjalainen et al. (1998) investigated D2 receptor density (B_max), affinity (K_d), and binding potential (B_max/K_d) using positron emission tomography (PET) and [¹¹C]raclopride (a D2 receptor radioligand) in healthy adult volunteers. Their results indicate that generally, women have lower D2 receptor affinity in the striatum compared to men. The lower D2 receptor affinity suggests an increase in endogenous striatal dopamine concentrations in women (Pohjalainen et al., 1998). Laakso et al. (2002) used PET and [¹⁸F]fluorodopa (an analog of L-DOPA which is the immediate precursor of dopamine) and confirmed that women have higher striatal dopamine synthesis capacity than men; furthermore, there was an effect of aging such that there was a decrease in dopamine activity with age in men but not women (Laakso et al., 2002). Taken together, these results suggest that males have higher D1 receptors in the nucleus accumbens (Andersen et al., 1997); women may have lower D2 receptor affinity compared to men (Pohjalainen et al., 1998), and this reduced D2 receptor affinity may result in higher striatal dopamine synthesis capacity in women relative to men (Laakso et al., 2002). These sexual dimorphisms in the structure and subsequent function of the MCLP may make men and women differentially susceptible to the reinforcing properties of rewards. To investigate this hypothesis directly, Munro et al. (2006) investigated sex differences in the magnitude of dopamine release, and subjective and neuroendocrine responses to amphetamine using PET and [¹¹C]raclopride. In stark contrast to the findings reported by Laakso et al. (2002), the results of the Munro et al. (2006) study revealed that men had greater dopamine release in the ventral striatum in response to amphetamines. Perhaps under baseline conditions, females have greater dopamine synthesis capacity (Laakso et al., 2002); however, when administered a rewarding substance, males have greater dopamine release in the ventral striatum (possibly a compensatory effect). Furthermore, Munro et al. (2006) report that the baseline binding potential of D2 receptors varies as a function of the menstrual cycle: women in the luteal phase had lower baseline binding potential in the putamen relative to women in the follicular phase (Munro et al., 2006). In the light of these findings, an investigation of hormonal influences on the MCLP is warranted.

To explore the possibility that the MCLP is modulated by gonadal hormones in humans, Dreher et al. (2007) obtained functional Magnetic Resonance Imaging (fMRI) scans from healthy women during the luteal and mid-follicular phases of their menstrual cycle. The fMRI paradigm included an event-related reward task; the results provide evidence of enhanced blood-oxygen-level-dependent (BOLD) responses associated with rewards during the mid-follicular phase of the menstrual cycle (when estrogen is not opposed by progesterone; Dreher et al., 2007). These results suggest that females respond more robustly to rewards during the phase of their menstrual cycle directly preceding ovulation. Furthermore, the results from the Dreher et al. (2007) neuroimaging study are in contrast to findings of hormonal modulation of impulsivity at the behavioral level. For example, Carroll et al. (2013) conducted an experiment that illustrates that rhesus monkeys self-administered higher levels of phencyclidine (PCP) during the luteal (vs. follicular) phase of the menstrual cycle and moreover, impulsive choice (on a delay-discounting measure administered to the same subjects) was greater during the luteal phase of the menstrual cycle. Smith et al. (2014) reported similar findings in a group of healthy adults who were administered a delay-discounting task. The results of their experiment revealed that the preference for an immediate reward declined during the middle of the menstrual cycle when fertility peaks. Similarly, Khaighobadi and Stevens (2013) administered the delay-discounting task to healthy females before and after exposure to either attractive male faces or neutral landscapes. The authors reported an interaction between fertility status and image type, such that women who were fertile and who had viewed images of attractive men chose the small, immediate reward less often than women at peak fertility viewing neutral images. The researchers conclude that women at peak fertility became less impulsive and that this was contradictory to their original hypothesis (Khaighobadi and Stevens, 2013). Interestingly, these results provide indirect evidence for the predictions proposed by Bjorklund and Kipp (1996). Due to the fluctuating levels of estrogen and progesterone across the menstrual cycle, females may adopt differential strategies to facilitate reproduction and mate selection during different phases of their menstrual cycle. For example, being impulsive might not incur enormous risks during phases of the menstrual cycle when females are not fertile—any risk incurred would only apply to the female herself and not to a developing embryo. In accordance with this hypothesis, Pine and Fletcher (2011) demonstrated that women in their luteal phase (non-fertile phase) are more impulsive than women in other phases of the menstrual cycle. However, according to parental investment theory, females must be selective in securing an appropriate mate and therefore, they may exhibit less impulsivity during periods of their menstrual cycle when they have the capacity to reproduce (Carroll et al., 2013; Khaighobadi and Stevens, 2013; Smith et al., 2014). These fluctuations as a result of hormonal changes during the menstrual cycle may help explain the mixed findings in delay-discounting in adult cohorts. For instance, the studies that indicate that women discount more steeply may have investigated women in non-fertile phases of the menstrual cycle. Conversely, the research that indicates that women are less impulsive may have inadvertently recruited females in the fertile phase of their menstrual cycle.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.