Folk psychological opinion views women as being more sensitive, emotional and even better mind readers as compared to men. In a scientific context these abilities refer to empathy and the adoption of a Theory-of-Mind (ToM). Empathy describes the sensitive perception of feelings and emotions of others , while Theory-of-Mind refers to the more cognitive aspect of inferring intentions, goals or desires of others . From an evolutionary perspective, having a ToM provides a powerful selective advantage, since it means the intentions of a counterpart can be evaluated in advance and responded to adequately. In highly developed societies appropriate social interactive behaviour in everyday situations (i.e. working environment, private and family life etc.) is regarded as a key cognitive ability and women are often regarded as superior in this respect.
In recent years, functional neuroimaging studies have increasingly addressed Theory-of-Mind related questions, including irony , jokes [4, 5], envy , false beliefs  and embarrassment . Other approaches utilize games and simulations that enable implicit and online ToM mechanisms to be investigated [9–16]. Here, the experimenter takes advantage of the distraction of the proper game, which puts the participants into a "flow-like" state and implicitly prevents the participant to direct their attention to the actual question of the taking of the perspective of a respective game partner on the basis of "I think that you think that I think... etc." Task designs applied to study implicit mentalizing comprise the Prisoner's Dilemma Game (PDG) [9, 11, 16, 17], the Ultimatum Game , economic decision games (Iowa Gambling Task) [15, 18] or a version of the stone-paper-scissor game .
The neural substrates associated with mentalizing comprise circumscribed regions of the right temporo-parietal junction (TPJ) and the medial prefrontal gyrus extending into the anterior (para)-cingulate cortex (ACC) [19–24]. The TPJ has been shown to be activated in relation to biological motion and to stimuli which signal intentions and intentional activity [22–25]. On the other hand, medial prefrontal and anterior cingulate regions are associated with planning and anticipation . Frith and Frith assume that in the context of mentalizing these regions play a major role in the anticipation of what and of how a partner is feeling, thinking or doing on the basis of what oneself would feel, think or do in the same situation [19, 27]. In turn, this shift permits the assessing and evaluating of ones own feelings and thoughts on a highly self-reflective level [28, 29].
Recent studies have shown that we also implicitly attribute intentions of various degrees to animals [30, 31] and to a somewhat lesser extent even to robots [9, 32] and computers . Neuroimaging findings indicate that humans do attribute self-generated actions, intentions and desires more greatly to human than to computer partners, though activity in the mentalizing network was detected in human-computer interactions as well. This was most evident in scenarios when the computer/machine was perceived as being directly responsive to the subjects' behavioural decisions [9, 11, 14, 17, 32].
The influence of gender on ToM along with its neural correlates has hardly been investigated. Accordingly, debates on this issue have been dominated by folk psychological positions and merely favour the point of view that women are superior in mentalizing and associated abilities.
In the present study we focused on the questions of whether humans attribute intentions and goals to human vs. computer partners and whether the participants' gender modifies neural correlates of implicit mentalizing. We applied a version of the PDG with subjects instructed to play either a putative human or a computer partner (while actually both were programmed to "play" a random sequence). By means of a cover story we were able to infer pure "intentional stance" [14, 33] associated neural activity as possible strategies of both partners were held constant between conditions. We hypothesized that subjects would attribute more real intentions and thoughts to the human partner, while the assumed random answer of the computer would elicit less brain activity in the medial prefrontal cortex. Regarding sex differences, we expected, in line with folk psychological assumptions that women would be better perspective takers and therefore display stronger signal changes in the medial prefrontal cortex.