Sometimes, we need to withhold the truth, whether for malicious or benign reasons. This act of withholding a true fact is termed “deception” 
. The involvement of brain processes in the manipulation of information suggests that brain activity during deception and truth telling should be different. It is this very assumption that founds the theoretical basis of neuroimaging studies on deception. Langleben et al. 
, Lee et al. 
, and Spence et al. 
were among the first to apply functional magnetic resonance imaging (fMRI) methodology to deception research. Using different experimental tasks in their studies of deception, they observed that activity in the prefrontal, cingulate, and parietal regions is associated with lying (increased brain activity during lying, relative to a truth-telling control). Other studies revealed that a number of these regions are implicated in a range of cognitive processes, such as executive control, working memory, attention, and inhibition 
. These results, which were corroborated in later functional neuroimaging research on deception 
, provided empirical evidence to support the insight from behavioral studies that deception is a complex and cognitively effortful task that demands a large amount of cognitive control and numerous mental management mechanisms 
The fMRI deception studies reviewed thus far offer insights into the brain activity associated with deception about affectively-neutral stimuli. However, little attention has been paid to the effect on brain activity when lying involves affective information. This represents a significant gap in the research, as the emotional attributes of suppressed information could have a very significant impact on deceptive responses. Recently, Abe 
employed positron emission tomography methodology to study the neural responses associated with deception in social interactions and reported activity in the medial orbitofrontal cortex and amygdala regions. Spence et al. 
elaborated on their previous studies 
and examined the neural activity associated with lying about past life events that the participants regarded as “embarrassing”. Lying was found to induce significant activity in the bilateral inferior frontal gyrus (Brodmann's Area [BA] 45 & 47); such activity tends to be associated with self-control and regulation 
. In these studies, the deceptions were low stake. Spence et al. 
reported the fMRI findings of a case study on deception involving a woman with possible Munchausen's syndrome by proxy who had been convicted of poisoning her child. They reported significant BOLD signals in the ventrolateral prefrontal and anterior cingulate regions. In this case study, both the content of the experimental materials used (i.e. whether she had poisoned a child) and the context of the deception (i.e. her wish to prove her innocence) could be emotion-provoking and high stake.
Indeed, on many social occasions, we may need to lie about the valence of a stimulus we encounter for reasons of social courtesy or in consideration of the feelings of others. Ganis et al. 
found that distinct neural networks support different types of deception, depending, for instance, on whether a lie is well-rehearsed or spontaneously made up. Furthermore, there is abundant evidence that emotional materials of different valence are processed differently in the brain 
. Mak et al. 
, for example, reported that the regulation of positive and negative emotions involves common as well as distinct neural correlates. Specifically, they observed that the regulation of positive emotions was associated with changes to the BOLD signals in the left dorsal prefrontal regions and in the left insula, amygdala, right rolandic operculum, and lingual gyri regions. In contrast, the regulation of negative emotions was associated with brain activity in the left orbitofrontal gyrus, left superior frontal gyrus, anterior cingulate gyrus, left middle occipital gyrus, and right precuneus regions.
Given all of the data reviewed, the study reported here examined low-stake deception. We investigated whether lying about the valence of affective stimuli involves activity in the frontal, cingulate, and parietal regions, as has been observed in deception studies using affectively neutral stimuli. Furthermore, we examined whether lying about the valence of stimuli involves distinct neural correlates, as has been observed in the study of emotion regulation. We employed the widely used International Affective Picture System (IAPS) as the experimental stimuli 
. The IAPS comprises a large set of standardized, emotionally evocative color photographs that are known to reliably induce emotional experiences. The stimulus set was tailored for each participant according to their rating of a set of affective pictures, which they selected from the IAPS, as either positive or negative. Orthogonal to this, the participants were cued to respond to each picture in either a truthful or a dishonest manner; for example, when cued to lie and presented with a negative picture, they had to respond that the picture was positive. Behaviorally, we would expect to perceive a typical “lie effect”, namely a significantly longer response time for the lie trials than for the truth trials. With regard to the neural correlates, we hypothesized that the regions associated with lying about positive and negative valence materials would be distinct from each other. Finally, based on the findings of the majority of imaging studies on deception, including our own work, we expected to find deception-related activations in the prefrontal cortex, anterior cingulate, and inferior parietal regions.