Reality = relevance? Insights from spontaneous modulations of the brain's default network when telling apart reality from fiction

Abstract

Background: Although human beings regularly experience fictional worlds through activities such as reading novels and watching movies, little is known about what mechanisms underlie our implicit knowledge of the distinction between reality and fiction. The first neuroimaging study to address this issue revealed that the mere exposure to contexts involving real entities compared to fictional characters led to engagement of regions in the anterior medial prefrontal and posterior cingulate cortices (amPFC, PCC). As these core regions of the brain's default network are involved during self-referential processing and autobiographical memory retrieval, it was hypothesized that real entities may be conceptually coded as being more personally relevant to us than fictional characters.

Methodology/principal findings: In the present functional magnetic resonance imaging (fMRI) study, we directly test the hypothesis that entity-associated personal relevance is the critical factor underlying the differential engagement of these brain regions by comparing the brain's response when processing contexts involving family or friends (high relevance), famous people (medium relevance), or fictional characters (low relevance). In line with predictions, a gradient pattern of activation was observed such that higher entity-associated personal relevance was associated with stronger activation in the amPFC and the PCC.

Conclusions/significance: The results of the study have several important implications. Firstly, they provide informed grounds for characterizing the dynamics of reality-fiction distinction. Secondly, they provide further insights into the functions of the amPFC and the PCC. Thirdly, in view of the current debate related to the functional relevance and specificity of brain's default network, they reveal a novel approach by which the functions of this network can be further explored.

Figure 1. Experimental Task and Design.

(a) Examples of scenarios, cues and correct responses to cues for all conditions. (b) A schematic representation of the sequence of events in a trial (trial length: 8 s). Across all experimental conditions, each trial began with a fixation cross (duration: 500 ms) which was followed by the presentation of single sentence for 2000 ms where a scenario was introduced. Following a delay (500 ms), a question cue was presented, to which the participant was required to respond. The cue remained on the computer screen for 1000 ms and the subject responded (yes or no) by pressing the appropriate button (index or middle finger) on a response box placed under the right hand. Variable jitter times were inserted before and after the scenario to enhance the temporal resolution of the blood oxygenation level-dependent (BOLD) signal. For the baseline rest period, a blank screen was presented for the duration of a trial.

Figure 2. Behavioral Findings.

The graphs display the results associated with each of the behavioral measures: reaction time (in milliseconds, top panel), response accuracy (in percentage of correct responses, middle panel) and perceived difficulty (in feedback ratings, bottom panel).

Figure 3. Gradient Relevance Pattern.

The top panel shows activations in and around the anterior mPFC and the PCC (x = −3) as a function of the indicated contrast. The left top panel shows the Friend>Famous contrast (inclusive mask: Friend>Control), the middle top panel shows the Friend>Fiction contrast (inclusive mask: Friend>Control) and the right top panel shows the Famous>Fiction contrast (inclusive mask: Famous>Control). All reported activations were corrected for multiple comparisons (p<.05) by employing cluster-size thresholding using Monte-Carlo simulations (initial height threshold: z = 3.09). Note that activations along the frontomedian wall extend into the dorsal medial PFC only in the Friend>Famous and Friend>Fiction contrasts. The bottom panel shows the averaged percentage signal change (PSC) response associated with all conditions within a peak voxel and its 26 adjacent neighboring voxels in the anterior medial prefrontal cortex (peak voxel: −5, 49, 0) (left bottom panel) and PCC (peak voxel: −5, −56, 30) (right bottom panel). The zero point in the graphs represents the resting baseline.

Figure 4. Other Relevance Patterns.

Each column shows activations resulting from the indicated contrast. The left column shows results from the Friend>Famous contrast (inclusive mask: Friend>Control), the middle column shows results from the Friend>Fiction contrast (inclusive mask: Friend>Control) and the right column shows results from the Famous>Fiction contrast (inclusive mask: Famous>Control). All reported activations were corrected for multiple comparisons (p<.05) by employing cluster-size thresholding using Monte-Carlo simulations (initial height threshold: z = 3.09). The top row shows the activation profile in the left lateral temporal gyri whereas the bottom row depicts the activation profile in the bilateral hippocampal formation across contrasts.