Introducing RISC: A New Video Inventory for Testing Social Perception
Bibliographic record
Abstract
Indirect forms of speech, such as sarcasm, jocularity (joking), and 'white lies' told to spare another's feelings, occur frequently in daily life and are a problem for many clinical populations. During social interactions, information about the literal or nonliteral meaning of a speaker unfolds simultaneously in several communication channels (e.g., linguistic, facial, vocal, and body cues); however, to date many studies have employed uni-modal stimuli, for example focusing only on the visual modality, limiting the generalizability of these results to everyday communication. Much of this research also neglects key factors for interpreting speaker intentions, such as verbal context and the relationship of social partners. Relational Inference in Social Communication (RISC) is a newly developed (English-language) database composed of short video vignettes depicting sincere, jocular, sarcastic, and white lie social exchanges between two people. Stimuli carefully manipulated the social relationship between communication partners (e.g., boss/employee, couple) and the availability of contextual cues (e.g. preceding conversations, physical objects) while controlling for major differences in the linguistic content of matched items. Here, we present initial perceptual validation data (N = 31) on a corpus of 920 items. Overall accuracy for identifying speaker intentions was above 80% correct and our results show that both relationship type and verbal context influence the categorization of literal and nonliteral interactions, underscoring the importance of these factors in research on speaker intentions. We believe that RISC will prove highly constructive as a tool in future research on social cognition, inter-personal communication, and the interpretation of speaker intentions in both healthy adults and clinical populations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".