His key insight was that we do not perceive the environment as such, but rather perceive it through its affor- dances, the possibilities for action it may provide. Fire, for example, affords warmth, illumination, and cooking, but at the same time it may also afford injury to the skin. What delineates warmth from injury is not always clear and depends on the species’ ability to detect the limit. What fire affords, then, is not merely a question of its physical properties, but its relation to a specific organism. As Gibson suggests, affordances ‘have to be measured relative to the animal’ (2015: 120). This does not mean, however, that affordances are indef- inite. For Gibson, affordances are invariant, meaning that they do ‘not change as the need of the observer changes. The observer may or may not perceive or attend to the affordance, according to his needs, but the affordance, being invariant, is always there to be perceived’ (2015: 130, emphasis in original). In its original meaning, an affordance is conceived as a relational property, one that ‘points both ways, to the environment and to the observer’ (Gibson, 2015: 121). The concept of affordance does not imply we can do anything with anything as ‘different layouts afford different behaviors for different animals [...] Knee-high for a child is not the same as knee-high for an adult, so the affordance is relative to the size of the individual’ (2015: 120). As Gibson put it, ‘affordances do not cause behavior but constrain and control it’ (1982: 411).

Hans centrale indsigt var, at vi ikke opfatter miljøet som sådan, men snarere opfatter det gennem dets affordancer, de handlemuligheder det kan give. Ild giver for eksempel varme, belysning og madlavning, men samtidig kan det også give skader på huden. Hvad der afgrænser varme fra skade er ikke altid klart og afhænger af artens evne til at opdage grænsen. Hvad ild giver, er altså ikke kun et spørgsmål om dens fysiske egenskaber, men dens relation til en bestemt organisme. Som Gibson foreslår, skal affordances 'måles i forhold til dyret' (2015: 120). Dette betyder dog ikke, at affordances er ubestemte. For Gibson er affordancer invariante, hvilket betyder, at de 'ikke ændrer sig, efterhånden som observatørens behov ændrer sig. Iagttageren kan eller kan ikke opfatte eller varetage affordancen, alt efter hans behov, men affordancen, der er invariant, er altid der for at blive opfattet' (2015: 130, fremhævelse i original). I sin oprindelige betydning er en affordance opfattet som en relationel egenskab, en der 'peger begge veje, til miljøet og til observatøren' (Gibson, 2015: 121).Begrebet affordance indebærer ikke, at vi kan gøre noget med noget, da 'forskellige layouts giver forskellige adfærd for forskellige dyr [...] Knæhøjt for et barn er ikke det samme som knæhøjt for en voksen, så råd er i forhold til individets størrelse« (2015: 120). Som Gibson udtrykte det, 'affordances forårsager ikke adfærd, men begrænser og kontrollerer den' (1982: 411).