Redoute proposed a model of sexual behavior, which differentiates four different components: a cognitive, an emotional, a motivational, and a physiological one [1
]. The cognitive component comprises an evaluation process during which the incentive value of a stimulus is appraised and attention is directed to it. The physiological component is related to autonomic and endocrinological changes, preparing the individual for sexual behavior in a physiological sense. The emotional component is seen as the subjective experience of hedonic feelings associated with sexual arousal, whereas the motivational component includes the desire to satisfy sexual arousal, resulting in goal-directed sexual behavior.
Over the last years, a growing number of studies have investigated brain processing of visual sexual stimuli with functional magnetic resonance imaging (fMRI) or Positron Emission Tomography (PET). These studies have found that erotic stimuli elicit activity in a widespread neural network, whose different parts have been associated with different components of Redoute's sexual behavior model [1
]. Cortical structures repeatedly found activated include parietal, occipitotemporal and frontal areas as well as the cerebellum, the insular cortex and the cingulate gyrus. Subcortical areas reported are the basal ganglia, the amygdala, the thalamus and hypothalamus (for an overview see Table ). High spatial resolution images acquired with 7 T fMRI even allow a clear functional distinction between adjacent subcortical structures such as the anterior caudate and mediodorsal thalamus [3
Potential substrates of erotic stimulus processing
Differences in stimulus induced brain activity reported by these studies might be related to stimulus content (sexual intensity: slightly erotic stimuli vs. pornographic material), presentation mode (static images vs. film sequences) and design type (presentation length of the stimulus material: blocked design vs. event-related design). The differences arising due to stimulus content and presentation mode have already been addressed by two imaging studies, one using videos and static stimuli of graded intensities [1
], the other comparing activation elicited by video clips that lead to sexual arousal and penile erection compared to static images that induced sexual arousal without erection [4
]. Whether emotion-inducing stimuli should be presented in a blocked or event-related fashion during fMRI is still a matter of debate. Especially in emotion research, the presentation duration appears to be important, not only from a methodological point of view, but also in respect of differences in information processing. Methodologically, blocked designs are superior in detecting activated brain areas, while event-related designs are superior in terms of estimation of the hemodynamic response function [5
]. Blocked designs are often used due to their easier implementation since randomization, jittering and spacing of different stimulus categories is not necessary. Furthermore, artefacts can be more easily detected by visual inspection of the data time course [6
]. Event-related designs are methodologically more demanding but allow for the randomization of stimulus presentation, thereby reducing confounds caused by stimulus order predictability [7
]. They further allow sorting trials post hoc according to behavioral responses, making this design more flexible than the blocked design [9
]. In respect of information processing, blocked designs differs from event-related designs. In blocked designs stimulus presentation length is prolonged and the occurrence of consecutive stimuli in a block is fully predictable [8
]. Therefore block designs might more readily activate cognitive processes such as sustained attention as well as top-down regulation mechanisms including inhibition of sexual arousal. This could decrease emotional involvement and hence change the underlying brain activation [11
Despite several excellent publications concerning theoretical methodological fMRI design considerations [5
] only a few attempts have been made to vary the design type while keeping the stimulus material and the experimental task constant. These studies compared blocked and event-related designs using different visual stimuli [15
], a finger opposition task [14
], auditory stimuli of different frequencies [16
] or higher cognitive tasks, such as semantic judgment [17
]. Only one study focused on emotional processing and to what extent brain activation elicited by stimuli depicting fearful and disgusting scenes depends upon the presentation design [11
]. In both conditions event-related in contrast to blocked stimulus presentation elicited greater activation in the insular and orbitofrontal cortex.
We know from everyday life that a split second is often enough to decide whether a person is attractive or a scene is erotic. Behavioral studies support this idea by showing that even sexual stimuli presented to the subconscious influence later processing [18
] or can be used as conditioned stimuli in a classical conditioning paradigm [20
]. Furthermore, correlates of erotic stimulus processing have been detected as early as 300 ms after stimulus presentation as shown in a recent electroencephalography (EEG) study by van Lankfeld et al. [21
Whether information processing is altered when erotic images are presented only for a brief period of time has not yet been addressed by existing imaging studies. A review of the literature on visually evoked sexual arousal revealed that fMRI studies have mainly used blocked stimulus presentation. None of these studies implemented an event-related design with brief presentation times of around one second.
In this study, we compared activation patterns elicited in an event-related design when erotic pictures are briefly (750 ms) presented with brain activity elicited by blocked stimulus presentation in which one block lasts around 20 s. We expected to identify an overlapping network for the blocked and the event-related design reflecting the neuroanatomical correlates of all four components of Redoute's [1
] neurobehavioral model of sexual arousal. We hypothesized that the blocked design is superior in identifying areas involved in cognitive aspects of erotic stimulus processing, such as sustained attention and inhibition of sexual arousal. Furthermore, we anticipated for the event-related design that activation would be more pronounced in structures associated with the motivational and emotional component of Redoute's [1
] neurobehavioral model, due to less influence from modulating cognitive processes.