Gibson and Gregory: Two Theories of Perception
Two Rival Theories of Perception
Psychologists disagree about how we turn raw visual information into a meaningful perception of the world. James Gibson argued that the environment itself provides everything we need to perceive it directly — no guesswork required. Richard Gregory argued the opposite: that perception is built up actively, using inference and stored knowledge. This disagreement maps onto psychology’s wider nature and nurture debate — Gibson’s theory sits on the nature side, Gregory’s on the nurture side.
Gibson’s direct theory of perception is a nature account: perception needs no learning, only the information already present in the visual world. Gregory’s constructivist theory of perception is a nurture account: perception is constructed using stored knowledge and experience.
Gibson’s Direct Theory of Perception
Gibson said that perception begins with sensory input from the eyes, but no further processing is needed to make sense of the world. As an observer moves through an environment, the visual information around them changes in a rich, structured pattern — texture becomes denser with distance, and the whole scene appears to flow around the point the person is heading towards. Gibson called this continuously changing pattern of visual information optic flow. He argued optic flow alone gives an observer everything they need to judge their speed, direction and the layout of the world around them.
Optic flow contains invariants — features of a scene that remain constant even though the observer is moving, such as the relative position of the horizon or the edges of a straight path. Because invariants do not change, Gibson argued they give direct, reliable information about the structure of the environment without any need for the brain to interpret or guess.
Gibson also argued that objects in the environment directly communicate their use through affordances. A flat, solid surface at knee height affords sitting; a handle-shaped object affords grasping. Because affordances are perceived directly from the shape and structure of an object, Gibson said no stored knowledge is needed to know what an object is for.
Gibson’s three key terms: optic flow (the changing pattern of visual information as an observer moves), invariants (features of the scene that stay constant during movement) and affordances (the clues an object gives about how it can be used). All three are picked up directly from the environment, with no inference needed.
Gregory’s Constructivist Theory of Perception
Gregory agreed that perception starts with sensory input, but argued this input is often incomplete, ambiguous or brief — too limited on its own to produce an accurate perception. To fill the gaps, Gregory said the brain forms a perceptual hypothesis: a best guess about what is out there, tested against incoming information much like a scientist tests a hypothesis against evidence.
Forming a perceptual hypothesis relies on inference — using existing knowledge to go beyond what is actually shown by the raw sensory data. This existing knowledge, built from past experience, is what Gregory called prior knowledge. Because perception depends on this stored knowledge, Gregory’s theory is described as top-down: what we already know shapes what we see, rather than the sensory input alone determining perception.
A strength of Gibson’s theory is that it explains the speed and accuracy of everyday perception. People judge distance, speed and the layout of a room almost instantly, which fits a theory in which perception happens directly from rich environmental information rather than through slow, effortful inference.
A weakness is that the theory struggles to explain visual illusions. If the environment always provides accurate, sufficient information, illusions such as ambiguous figures should not be able to fool the visual system. The existence of illusions suggests the brain does more than passively pick up information — a problem Gregory’s theory does not have.
A strength of Gregory’s theory is that it explains visual illusions well. If perception involves forming a hypothesis based on sensory input and prior knowledge, then illusions occur when that hypothesis is wrong — for example when the brain misapplies size or depth cues learned from a lifetime of viewing flat surfaces and corners.
A weakness is that the theory may over-state how much inference is needed for simple, familiar scenes. Gibson’s research on skilled real-world tasks, such as pilots judging their approach to a runway, shows perception can be extremely fast and accurate without an obvious step-by-step process of hypothesis-testing.
Haber and Levin (2001): Size and Distance Perception
Haber and Levin (2001) tested whether size perception and distance perception rely on the same source of information. Participants viewed a range of objects placed at different distances outdoors — some highly familiar, with a well-known real-world size, such as a basketball or a chair, and others unfamiliar, with no fixed known size. For each object, participants estimated both its size and how far away it was.
The results showed that size judgements for familiar objects were accurate and depended mainly on participants’ stored knowledge of the object’s real-world size, largely independent of the depth cues available in the scene. Size judgements were most accurate for objects with little natural size variation between examples (token-invariant objects), and distance information had little or no effect on size estimates; accuracy for size depended on the object’s consistency of real-world size rather than on distance cues. Distance judgements were most accurate for familiar, standard-size objects (e.g. a door or milk bottle) and less accurate for objects of variable size and for unfamiliar cut-out shapes, because participants used their stored knowledge of familiar objects’ size to judge how far away they were.
Haber and Levin concluded that size perception and distance perception are separate, independent processes. Judging the size of a familiar object draws on prior knowledge stored in memory — evidence for Gregory’s constructivist idea that perception uses stored knowledge. Judging the distance of familiar objects also drew on stored knowledge of their real-world size, so participants were more accurate with familiar than unfamiliar objects — evidence that fits Gregory’s constructivist theory rather than Gibson’s direct theory. The study therefore provides strong support for Gregory’s theory across both size and distance perception.
Haber and Levin (2001) found that both size and distance judgements were more accurate for familiar, standard-size objects because participants relied on stored knowledge of those objects’ real size (supporting Gregory). Their conclusion was that size and distance perception are independent processes, but both showed the influence of prior knowledge on familiar objects.
A strength of the study is its ecological validity. Real objects were viewed outdoors at genuine distances, rather than through artificial images in a laboratory, so the findings are likely to reflect how size and distance perception actually work in everyday life.
A weakness is that the task relied on participants giving verbal or numerical estimates of size and distance, which is a subjective measure — two people might interpret how far away something is differently, reducing the reliability of the results. The sample was small and unrepresentative — only 9 male US college students with good eyesight — which limits how confidently the findings can be generalised to the wider population (e.g. women, other age groups, or people with poorer vision).