Psych Perception Memory Learning Cognition And Emotion

---
name: psych-perception-memory-learning-cognition-and-emotion
description: "Use when working with the well-established core of cognitive psychology: perception and attention including change blindness and attentional limits, memory and its systems, reconstruction and the reliability of eyewitness recall, learning and conditioning, cognition, reasoning and the biases that replicated versus those that did not, and emotion and the competing theories of what an emotion is."
---

# Psychology, Sociology and Cultural Sciences: Perception, Memory, Learning, Cognition and Bias, and Emotion

> **Part 2 of 5** of the *Psychology, Sociology and Cultural Sciences* reference (plugin `psychology-sociology-cultural-sciences`), covering §4–§8. Sibling skills: `psych-epistemic-situation-and-replication-reform` (§0–§3), `psych-development-personality-social-clinical-and-neurodiversity` (§9–§15), `psych-sociology-institutions-culture-and-the-weird-problem` (§16–§24), `psych-reference` (§25–§29). Section numbers are shared across the set; a reference written as §N → `skill` points into that sibling skill.
>
> **Currency:** The core findings are decades stable. Two areas are live. See §25 → `psych-reference` for where replication reform actually stands, and the social-media-and-adolescent-mental-health dispute presented as the dispute it is.

> **⚠️ Read §1 → `psych-epistemic-situation-and-replication-reform` first. It is not preamble — it is the single most useful thing in this
> document.**
>
> **In most fields you can learn the findings and treat the epistemics as background.**
> ⚠️ **Here that gets you a head full of confident claims that are false.** **Between
> roughly 2011 and 2020 these fields discovered that a large fraction of their published
> literature did not replicate**, and ⚠️ **much of the popular material about psychology
> was written from the pre-crisis literature and is still in circulation — in
> bestsellers, TED talks, management training and journalism.**
>
> **⚠️ GOTCHA** boxes mark claims that are widely repeated and either false, much weaker
> than advertised, or genuinely contested.
>
> **The three ideas that organize everything below:**
> 1. **⚠️ Knowing WHICH findings survived is more valuable than knowing more findings**
>    (§1–§3 → `psych-epistemic-situation-and-replication-reform`).
> 2. **⚠️ The robust results cluster in a recognizable place**: **perception,
>    psychophysics, memory, learning, individual differences, and large-sample
>    demography.** **The fragile ones cluster in social priming, small-N lab studies with
>    surprising outcomes, and anything that made a good headline** (§1.3 → `psych-epistemic-situation-and-replication-reform`).
> 3. **⚠️ Humans are far more shaped by situation and structure than intuition allows —
>    and far less infinitely malleable than a century of blank-slate theorizing
>    assumed.** **Both errors are common, and they're symmetrical.**

---

## §4. Perception and Attention

**⚠️ The most robust area in psychology, and the least publicly known.** **Psychophysics
has replicable, quantitative laws** — **Weber-Fechner, Stevens' power law, signal
detection theory** (⚠️ **which cleanly separates sensitivity from response bias and is
one of psychology's genuinely great ideas**).
**Perception is constructive, not receptive**: **the visual system infers a stable world
from a degraded, saccade-interrupted, two-dimensional input** — **which is why illusions
work and why they're informative rather than mere curiosities.**
**Attention as selection**: ⚠️ **inattentional blindness and change blindness are robust
and genuinely startling** — **people miss large, obvious events when attending
elsewhere.** ⚠️ **The practical implication is that eyewitness confidence about what was
in view is much weaker evidence than it feels** (§5).
**⚠️ Multitasking is task-switching with a cost**, and **the cost is reliably
underestimated by the person paying it.**

---

## §5. Memory

**⚠️ Memory is reconstructive, not reproductive — and this is one of the most robust and
most socially consequential findings in psychology.**
**⚠️ You do not retrieve a recording. You reconstruct an account from partial traces,
schemas and current context** — **and the reconstruction feels identical to a recording
from the inside.**

```
Encoding → Storage → Retrieval, with ⚠️ distortion possible at every stage
Working memory   ⚠️ limited (the famous "7±2" is now generally considered closer
                 to ~4 chunks); Baddeley's components
Long-term        declarative (episodic/semantic) vs procedural
⚠️ Reconsolidation  retrieving a memory makes it labile — recall can CHANGE it
```
> **⚠️ GOTCHA — false memories are easy to create and indistinguishable from true ones to
> the person holding them.** **Loftus's work is among the most replicated in the field:
> leading questions alter recall, and entire plausible events can be implanted in a
> meaningful minority of participants.** ⚠️ **Confidence does not track accuracy well —
> especially after a witness has been given confirming feedback.** **This has direct
> consequences for eyewitness testimony, investigative interviewing, and therapeutic
> "recovered memory," where the harm has been documented and severe.**

**⚠️ What actually improves learning** — **and this is robust, practically useful, and
routinely ignored:**
- **⚠️ RETRIEVAL PRACTICE (testing yourself) beats re-reading substantially.**
- **⚠️ SPACED practice beats massed practice (cramming) for retention.**
- **Interleaving beats blocking for discriminable skills.**
- **Elaboration and self-explanation beat highlighting.**
- ⚠️ **The techniques that FEEL most effective — re-reading, highlighting, massed
  practice — are the least effective.** **Fluency is mistaken for learning, which is why
  students reliably choose the worse strategy.**

---

## §6. Learning and Conditioning

**Classical conditioning** (Pavlov) — ⚠️ **and the modern understanding is informational,
not reflexive: the Rescorla-Wagner model says learning tracks PREDICTION ERROR, which
turned out to map onto dopamine signalling and became the foundation of reinforcement
learning in AI.**
**Operant conditioning** (Skinner) — **reinforcement and punishment, positive and
negative.** ⚠️ **Schedules matter enormously: VARIABLE-RATIO reinforcement produces the
most persistent, extinction-resistant behaviour**, which is why slot machines, loot boxes
and infinite feeds are built the way they are (see a trading reference §16).
**⚠️ Extinction is not erasure** — **spontaneous recovery and renewal show the original
learning is retained and suppressed, which matters for therapy and relapse** (§13 → `psych-development-personality-social-clinical-and-neurodiversity`).
**Observational learning** (Bandura), **habit formation** (⚠️ **cue-routine-reward; and
the widely cited "21 days" figure has no good basis — the actual research found a wide
range averaging far longer**).

---

## §7. Cognition, Reasoning and Bias

**⚠️ Dual-process framing (System 1/System 2) is a useful teaching heuristic and should
not be reified.** ⚠️ **There are not two systems in the brain; there is a spectrum of
automaticity, and the strong two-system claim has been substantially criticized.**

**⚠️ Heuristics and biases — handle with calibration.** **The core Kahneman-Tversky
findings are real and important: availability, representativeness, anchoring, framing,
loss aversion, the conjunction fallacy.** ⚠️ **What has NOT held up well is the more
elaborate downstream literature, several priming-adjacent effects, and the effect sizes
claimed in popular treatments.** **Kahneman himself publicly acknowledged that the priming
chapter of *Thinking, Fast and Slow* rested on evidence that did not hold.**
> **⚠️ GOTCHA — anchoring is robust; ego depletion is not.** ⚠️ **Ego depletion (willpower
> as a depletable resource, the glucose story) was enormously influential, spawned a large
> literature, and has substantially failed large multi-lab replication.** **If you learned
> self-control from pre-2015 popular material, you learned this.**

**⚠️ The Gigerenzer critique deserves its place**: **many "biases" are ecologically
rational — fast and frugal heuristics that perform well in the environments people
actually inhabit** — **and framing problems in natural frequencies rather than
probabilities dramatically improves performance, including among physicians.**
**Bayesian reasoning, base rate neglect, and ⚠️ the fact that most people find
probabilistic reasoning genuinely difficult** — **which is a design problem as much as a
cognitive one.**
**⚠️ Motivated reasoning and identity-protective cognition are robust and depressing**:
**greater numeracy and knowledge can INCREASE polarization on identity-linked issues,
because more capable reasoners are better at defending prior commitments.**

---

## §8. Emotion

**⚠️ The basic-emotions view (Ekman — a small set of universal, discretely expressed
emotions) is genuinely contested**, and **the constructionist view (Barrett) holds that
emotions are constructed from core affect plus conceptual knowledge, with far more
cultural and individual variation in expression and recognition than the universality
claim allows.** ⚠️ **Treat "the six universal facial expressions" as a live dispute, not
settled fact.**
**Appraisal theory**: **emotion follows from evaluation of a situation relative to goals**
— **which is the theoretical basis of cognitive therapy** (§13 → `psych-development-personality-social-clinical-and-neurodiversity`).
**⚠️ The facial feedback hypothesis in its strong form (holding a pen in your teeth makes
cartoons funnier) failed a large multi-lab replication**, though weaker versions retain
some support.
**Affective forecasting** — ⚠️ **people systematically overestimate the intensity and
duration of future emotional reactions, and underestimate their own adaptation.** **This
is robust and practically useful.**
**Emotion regulation**: ⚠️ **reappraisal generally outperforms suppression** — **suppression
is effortful, reduces expression without reducing experience, and carries social costs.**