[Best I have found for a good discussoin of how ‘affordances’ were intended compared to the common use]

BI 223 Vicente Raja: Ecological Psychology Motifs in Neuroscience

BI 223 Vicente Raja: Ecological Psychology Motifs in Neuroscience | Brain Inspired

BI 223 Vicente Raja: Ecological Psychology Motifs in Neuroscience

---

Here is a ChatGPT summary of the content and core points

Podcast summary

Vicente Raja is arguing for an ecological neuroscience: a neuroscience that takes body, movement, and environment as constitutive, not decorative. The brain shouldn’t be studied as if cognition happens in a skull-bound computer, then gets applied to the world. Cognition happens through organism-environment coupling.

His orthodox Gibsonian claim is that affordances are perceived through ecological information. An affordance isn’t just ‘a possible behaviour’. It’s a property of the environment relative to an organism, specified by information in stimulation. Perceiving it means detecting that information.

He accepts that ‘affordance’ has escaped ecological psychology. In design, robotics, and neuroscience, it now often just means ‘opportunity for interaction’. Don Norman’s door example is the clean case: don’t put a pull-handle on a door that should be pushed. But Raja says that’s already no longer Gibsonian in the strict sense.

A large chunk is about ‘motifs’. Raja’s idea is that scientific fields rely on vague but productive core concepts: ‘representation’, ‘encoding’, ‘affordance’, ‘algorithm’, and so on. They aren’t precise definitions. They are flexible explanatory patterns. Their vagueness is partly why they survive and travel between fields. Annoying, but useful.

He contrasts mainstream cognitive neuroscience with ecological psychology. Mainstream models often assume poor stimulation plus internal inference: the brain adds prior knowledge to make perception work. Ecological psychology starts from the opposite bet: the stimulus is rich, and the task is to identify the information already available in the organism-environment relation.

His radio metaphor matters. The radio doesn’t reconstruct the song from scratch; the information is in the signal. The organism still does work, but not the same kind of work. The brain doesn’t have to rebuild the world internally every moment. It helps couple movement to meaningful information.

He pushes neuroscientists to take the stimulus seriously. Don’t just stick people in fMRI scanners and show them pictures, then assume you’ve studied perception. Real objects, real movement, and naturalistic environments matter. He cites work where real objects and pictures produce different effects.

Resonance is his preferred bridge concept. Ecological information constrains organism-environment dynamics, and may also constrain neural dynamics. So the brain is not representing the world in the usual computational sense; it is resonating with, or being tuned to, meaningful environmental structure.

He doesn’t think ecological psychology explains everything. It’s strong for perception-action in natural environments. Scaling it up to reading, language, social cognition, and richer mental life is still a problem. He’s wary of ecological psychology becoming a theory of everything. Good. The world has enough imperial theories wearing a borrowed hat.

The plant behaviour section extends the argument. Plants move. Climbing plants change their movement patterns when a support pole is nearby, even before touching it. Raja’s lab is studying whether this resembles goal-directed behaviour, and whether plants are detecting environmental possibilities for action in a minimal sense.

Affordances in brief

Colloquially, ‘affordance’ usually means ‘what something lets you do’. A handle affords pulling. A button affords pressing. A website affords sharing. This is the Norman/design use. Useful, but flattened.

Properly, in the ecological sense Raja defends, an affordance is not just a feature or option. It is an organism-relative possibility for action, specified by ecological information. The key triad is:

environment,
organism,
information specifying possible action.

So: a pole affords climbing for a climbing plant, not for a toaster. A chair affords sitting for some bodies, not all bodies. A handle may afford pulling, but only within a learned, embodied, practical world.

The mistake is to treat affordances as object properties. Better: affordances are relational, embodied, and perceptible action-possibilities in a structured environment.