Picture perfect peaks: comprehension of inferential techniques in visual narratives

Abstract

The ability to reconstruct a missing event to create a coherent interpretation – bridging inference – is central to understanding both real-world events and visual narratives like comics. Most previous work on visual narrative inferencing has focused on fully omitted events, yet few have compared inference generation when climactic events become replaced with a panel employing numerous inferential techniques (e.g., action stars or onomatopoeia). These techniques implicitly express the unseen event while balancing several underlying features that describe their informativeness. Here, we examine whether processing and inference resolution differ across inferential techniques in two self-paced reading experiments. Experiment 1 directly compared five distinct types, and Experiment 2 explored the effect of combining techniques. In both experiments, differences in processing arise both between inferential techniques themselves, and at subsequent panels allowing the bridging inference to be resolved. Analysis of inferential features suggested that the explicitness of the inferential technique led to greater demand in processing, which later facilitated inference generation and comprehensibility. The findings reinforce the necessity of discussing the diversity of narrative patterns motivating bridging inferences within visual narratives.

Association of Autistic Traits with Inference Generation in Visual Narratives

“…Indeed, the results from the ACC analyses suggested that participants with lower levels of autistic traits showed greater ACC, as well as longer RTs. Although we did not nd evidence of a speed-accuracy trade-off in our data, this proposal is in line with prior ndings that longer RTs are often associated with higher VLFI scores [23]. Interestingly, this performance modulation only occurs for those with lower levels of autistic traits; for those with higher levels of autistic traits, visual language uency did not modulate either ACC or RT.…”

“…With regard to RT, we found that for those with lower levels of autistic traits, increasing visual language uency was associated with increased RTs for panel deletion identi cation. This nding suggests that having higher rates of experience with visual narratives may negatively in uence one's ability to determine where a panel was missing from, which is in line with prior inferencing research (see [23]).…”

“…For Experiment 1, we asked participants to identify the position of a missing panel in a visual narrative. We expected participants to have an easier time recognizing when the Peak panel was removed, since it is the narrative category most associated with facilitating inference generation [20], [21], [23], [31]. Indeed, we found that the highest mean ACC for recognizing deleted panels was for Peaks, as compared to Initials and Releases.…”

“…We used a procedure similar to REMOVED FOR BLIND COPY (2014)'s Experiment 4 [31] (see also [23], in which participants identi ed where a panel had been deleted from in a comic strip. Participants were shown a 5-panel comic strip with numbered arrows pointing to the spaces before and after each panel, and were told, "We've deleted a panel from this comic.…”

“…REMOVED FOR BLIND COPY (2019) [21] highlighted a number of mechanisms that could require individuals to draw inferences in visual narratives, including panel deletions (i.e., removing a panel: a technique utilized in the current studies); panel replacements (e.g., replacing a panel with an "action star," a tool commonly used to depict events and sounds [typically when accompanied by onomatopoetic text] in a narrative; [20], [20]; and onlookers (e.g., diverting the focus of events in a narrative by, for example, having a character in frame viewing an event out of frame). Klomberg and REMOVED FOR BLIND COPY (2022) [23] found that the elements of explicitness and framing (i.e., "windowing part of the actual scene" [p. 599]; [23]) of panels were the most in uential features affecting narrative comprehensibility and processing.…”