MGMT20001 · Organisational Behaviour
Perception and Attribution
Perception is the process of selecting, organising and interpreting sensory data to make sense of the world — and it is perception, not objective reality, that drives behaviour. Because the process takes shortcuts it produces systematic errors (stereotyping, halo, primacy/recency, projection, self-fulfilling prophecy); those feed flawed attributions — the causal stories we tell about why people act as they do, skewed by the fundamental attribution error, self-serving bias and the actor–observer effect. The same shortcuts bias decisions: the rational model is an ideal, but real deciders satisfice under bounded rationality (Simon) and fall to biases (anchoring, confirmation, framing, escalation, Dunning–Kruger). The Women-in-Engineering mini-case threads the whole chain — perception → attribution → decision → intervention.
What this chapter covers
- 01The perceptual process: select → organise → interpret
- 02Common perceptual errors (stereotyping, halo, primacy/recency, projection, self-fulfilling prophecy)
- 03Attribution theory: internal vs external causes
- 04The three attribution biases (FAE, self-serving, actor–observer)
- 05Decision-making: the rational model vs bounded rationality & satisficing (Simon)
- 06The decision-bias zoo (anchoring, confirmation, framing, escalation, Dunning–Kruger)
- 07Case: Women in Engineering — perception → attribution → decision → intervention
Worked example: run the chain on a hiring scenario
- +1Perception: filing her as 'not a typical engineer' is stereotyping driven by selective attention / unconscious bias.
- +1Attribution: crediting her wins to luck/help (and a man's to talent) is the fundamental attribution error — a biased causal story.
- +1Decision: hiring the 'good-enough', familiar-fitting candidate is satisficing plus confirmation bias; feeling 'confirmed' afterwards is the self-fulfilling loop.
- +1Recommend a design fix: you can't will bias away — design it out: structured criteria set before seeing candidates, blind/anonymised screening, diverse panels.
- +1Seek disconfirming evidence: explicitly look for data that would overturn the favoured choice, to fight confirmation bias.
Key terms
- Perceptual process
- Select → organise → interpret: we attend to a fraction of stimuli, sort them into categories, then assign meaning. We act on the perceived world, not the objective one.
- Self-fulfilling prophecy (Pygmalion)
- A causal loop, not a one-off misread: the perceiver's expectation changes how they treat the target, which changes the target's behaviour, which 'confirms' the expectation.
- Fundamental attribution error (FAE)
- When judging others, over-weighting their disposition (character) and under-weighting the situation. Distinct from self-serving bias, which is about judging ourselves favourably.
- Bounded rationality (Simon)
- Real deciders have limited information, time and processing, so they satisfice — take the first 'good enough' option rather than the optimal one. Alongside it, intuition is fast pattern-matching from experience.
- Escalation of commitment
- Throwing more resources at a failing course to justify the past — not persistence. The tell: the future case for the project no longer stands on its own, only the sunk past does.
Perception and Attribution FAQ
What's the difference between the FAE and self-serving bias?
Direction of judgement. The fundamental attribution error is about judging others (we over-blame their character and under-weight the situation). Self-serving bias is about judging ourselves (we take credit for wins and blame the situation for losses). The actor–observer effect is the combined asymmetry.
Is the rational decision model what people actually do?
No — it's the ideal (identify the problem, weight criteria, generate and evaluate alternatives, choose, review). It assumes full information and unlimited processing, which no one has. Simon's bounded rationality is the realistic correction: we satisfice.
How is the self-fulfilling prophecy more than a misjudgement?
It is causal. The perceiver's expectation changes how they treat the target (fewer chances for a low-rated hire), which changes the target's behaviour (underperformance), which appears to 'confirm' the original expectation. Scenarios describing exactly this are testing the Pygmalion effect.
How do I fix bias in a perception/decision scenario?
Design it out rather than 'trying to be less biased': structured, criteria-based decisions set before seeing candidates; blind or anonymised screening; diverse panels; and a deliberate search for disconfirming evidence. That earns the application marks.
Exam move
Treat this as one chain, not three lists: a perceptual error feeds a biased attribution which drives a flawed decision — so structure scenario answers as perception → attribution → decision → intervention, naming the specific error at each step. Keep the FAE vs self-serving split crisp (others vs self). For decisions, contrast the rational ideal with bounded rationality / satisficing (Simon) and be able to name a bias on sight. Then always finish with a design fix (structure, blinding, criteria), which is where the application marks live.
