Welcome to CogSci Unpacked, an exciting blog series dedicated to summarizing academic papers from the Cognitive Science, a CSS Journal. Our goal is to bridge the gap between academia and the broader public, fostering a better understanding of cognitive science and making it accessible and relatable to all. If you’re curious to dive even deeper, we invite you to explore the full academic paper.
Why do humans gesture when they speak? It sounds like a simple question, yet cognitive science has never quite settled on an answer. Are gestures deliberate communicative tools shaped by an audience? Or are they largely unconscious by-products of thinking and speaking?
Part of the difficulty lies in what researchers have traditionally counted as a gesture. Most studies focus on representational movements – iconic gestures that visually depict meaning, like tracing a spiral in the air to describe a winding road, or shaping the hands to show the size of an object. But everyday conversation contains another, less discussed class of movements. Imagine someone saying, “You know what I mean?”, while extending an open palm toward their interlocutor. The gesture does not depict an object or action. Instead, it manages the interaction itself – inviting agreement, marking shared understanding, or softening a claim.
These are interactive gestures, and they were the unexpected starting point of our recent study.
We recorded participants engaged in extended, face-to-face conversations under varying visual and conversational conditions. Initially, gestures were annotated using standard categories common in gesture research: iconic, metaphoric, deictic (pointing), emblematic (conventional signs like thumbs-up), pantomimic, and beat gestures. Surprisingly, nearly a quarter of all gestures resisted classification. Rather than treating these movements as noise, a second analysis asked a simple question: were they interactive?
They overwhelmingly were. Almost 90% of gestures that do not fall into traditional categories were found to play an interactive role, accounting for over a quarter of the entire dataset. In frequency terms, interactive gestures were not marginal phenomena but central components of conversational behaviour – at least as prevalent as many gesture types that dominate the literature.
This finding alone invites a shift in perspective. If cognitive science aims to understand how communication works in real time, movements that regulate shared understanding may be as theoretically important as those that visually depict meaning.
The study’s central manipulation examined gesture visibility. In one phase, speakers and listeners could see one another normally. In another, a screen occluded visibility of the torso and hands while preserving access to the face. The logic was straightforward: if gestures primarily serve communicative functions, blocking visibility should suppress gesture production.
Interactive gestures did decrease when visibility was blocked, but only in simple conversations.
When discussions involved more complex topics (those involving emotionally and socially challenging concepts) in this visual occlusion condition, participants’ interactive gesture rates remained stable. Participants continued to produce discourse-managing, audience-directed gestures even when those gestures could not be seen, only when the conversations were complex.
This selective dissociation poses a challenge for familiar theoretical positions. A strictly pragmatic account predicts broad suppression under occlusion: why deploy invisible communicative signals? A strictly unconscious account predicts minimal sensitivity to visibility: why should occlusion matter at all?
Instead, gesture behaviour appears shaped by both communicative context and conversational demands.
One interpretation is that gesture operates at a level best described as subconscious. On this view, gesturing reflects intrinsic social pressures within the cognitive system – pressures tied to being polite tied to intersubjective acknowledgement – that are neither fully deliberate nor purely automatic. Visual feedback modulates these pressures, but does not fully determine them. Under sufficient task difficulty, arising as social-emotional complexity in conversations, the drive to maintain interactional coherence may override visibility constraints entirely.
One of the most consequential implications of this work is methodological. Gesture research has historically relied on constrained tasks and narrative paradigms, which historically yield detailed analyses only of representational gestures. Observing behaviour in sustained, semi-naturalistic conversation reveals a different distribution of gesture types, highlighting movements that regulate interaction rather than depict content.
Rather than viewing gestures as optional accompaniments to language, the present findings point toward something more fundamental. Gesturing may be tied to a subconscious drive to express oneself. This drive only becomes experimentally observable in interaction, and it appears shaped both by the emotions arising in conversation and modulated by interlocutor visibility, but not fully dependent on either. The persistence of invisible gestures highlights a cognitive system organised around expression as much as communication. For cognitive science, this reframes gesture as evidence of how thought, action, and social engagement remain deeply intertwined.
Facts Only
Researchers conducted a study on human gestures during face-to-face conversations.
Participants engaged in extended conversations under varying visual and conversational conditions.
Gestures were initially categorized using standard types: iconic, metaphoric, deictic, emblematic, pantomimic, and beat gestures.
Nearly 25% of observed gestures did not fit traditional categories.
A second analysis revealed that 90% of unclassified gestures were interactive, managing conversation flow.
Interactive gestures accounted for over 25% of all gestures in the dataset.
The study manipulated gesture visibility by occluding participants' hands and torsos while keeping faces visible.
In simple conversations, interactive gestures decreased when visibility was blocked.
In complex conversations (emotionally or socially challenging topics), interactive gesture rates remained stable even when unseen.
The study suggests gestures may reflect a subconscious drive tied to social and emotional pressures.
Traditional gesture research has focused on constrained tasks, potentially overlooking interactive gestures.
The findings challenge purely pragmatic or unconscious explanations for gesturing.
Executive Summary
A recent study in cognitive science challenges traditional views on why humans gesture during speech. Researchers recorded participants in face-to-face conversations under varying conditions, initially categorizing gestures into standard types like iconic, metaphoric, and deictic. However, nearly a quarter of gestures didn’t fit these categories. Upon closer analysis, 90% of these unclassified gestures were found to be interactive—movements that manage conversation flow, such as inviting agreement or softening claims, rather than depicting content. These interactive gestures accounted for over a quarter of all observed gestures, suggesting they are as fundamental to communication as representational gestures.
The study further manipulated gesture visibility by occluding participants' hands and torsos while preserving facial visibility. In simple conversations, interactive gestures decreased when visibility was blocked, but in complex discussions involving emotionally or socially challenging topics, gesture rates remained stable even when unseen. This suggests that while visual feedback influences gesture production, the drive to maintain interactional coherence can override visibility constraints in demanding contexts. The findings imply that gestures may reflect a subconscious social-cognitive drive, shaped by both communicative needs and emotional complexity, rather than being purely deliberate or automatic.
Full Take
This study offers a compelling reframing of human gesturing, shifting focus from representational movements to the often-overlooked role of interactive gestures. The strongest version of this narrative is its evidence-based challenge to binary explanations—gestures are neither purely deliberate communicative tools nor unconscious byproducts of thought. Instead, they appear to operate at a subconscious level, shaped by social pressures, emotional complexity, and the need for interactional coherence. The persistence of unseen gestures in complex conversations suggests a deeper cognitive drive to express and connect, even when visual feedback is absent.
Pattern scan: The article avoids manipulation tactics, presenting findings with nuance and acknowledging uncertainty. It resists oversimplification, instead highlighting the complexity of gesture production. No patterns of distortion, bad faith, or emotional exploitation are detected. The discussion of methodological limitations (e.g., historical focus on representational gestures) demonstrates intellectual honesty.
Root cause: The narrative reflects a broader paradigm shift in cognitive science toward embodied and socially situated cognition. It challenges the assumption that gestures are secondary to language, instead positioning them as integral to thought and interaction. This aligns with growing recognition of the body’s role in cognition, echoing theories like embodied cognition and distributed cognition.
Implications: For human agency, this suggests that communication is not just about transmitting information but about co-constructing understanding through subtle, often subconscious social cues. The findings could inform fields like education, therapy, and human-computer interaction, where understanding nonverbal behavior is critical. However, the study’s reliance on semi-naturalistic conversations may limit generalizability to other contexts.
Bridge questions: How might these findings apply to digital communication, where gestures are absent or mediated? Could interactive gestures be culturally specific, and if so, how? What would it mean for AI or robotic systems to incorporate such subconscious social signaling?
Counterstrike scan: A bad actor pushing this narrative might exaggerate the findings to argue that gestures are entirely subconscious, undermining the role of deliberate communication. However, the actual content resists this binary, emphasizing the interplay between conscious and unconscious processes. No structural alignment with manipulation tactics is detected.
Patterns detected: none