When Expectations Shape What We Think We See
Two people can encounter the same evidence and construct different explanations. Understanding that difference requires attention to what they saw, what they expected and what the material actually established.
Imagine watching a short interview after being told that the speaker is highly experienced. A hesitant answer might seem thoughtful. Now imagine encountering the same footage after being told that the person is poorly prepared. The hesitation might appear to confirm a lack of competence.
Neither interpretation necessarily follows from the pause alone. Yet the information supplied before a recording can influence which details seem important afterwards.
This raises a question central to media and communication: when we believe a recording confirms an explanation, how much of that explanation did we bring to the recording ourselves?
What the experiment actually showed
In January 1983, psychologists John Darley and Paget Gross published an experiment involving 67 analysed undergraduate participants. They investigated whether information about a child's socioeconomic background would affect assessments of her academic abilities.
Participants received material suggesting either a relatively affluent or disadvantaged background. Some then viewed the same videotape of a child, Hannah, answering academic questions. Her performance was deliberately mixed: she answered some easy and difficult questions correctly and others incorrectly.
When participants had received background information but no performance video, their assessments were comparatively restrained. After viewing the mixed performance, those given the affluent-background information rated Hannah's ability more favourably than those given the disadvantaged-background information. Both groups referred to the recorded performance to justify their conclusions.
The experiment demonstrated different judgements under experimentally varied contextual information. The researchers interpreted this as hypothesis-confirming assessment. It did not establish Hannah's actual academic ability, nor does it show that all audiences interpret every recording in this way.
When an inference becomes an apparent fact
A label or introductory description can establish an expectation before the audience encounters the main evidence. If the subsequent material is ambiguous, individual details may be interpreted as confirmation of that expectation.
This is relevant to news, documentaries and reality television, where audiences often encounter people through a mixture of narrated descriptions, recorded behaviour and editing. But the experiment concerned one educational judgement, not a representative sample of media programmes. Applying its findings to audiovisual communication is a reasoned extension, not an additional experimental result.
An important alternative
Background information is not inherently misleading. A person's circumstances, relevant experience or the conditions in which an event occurred may be necessary for fair interpretation. Ignoring context can be as unhelpful as relying on it unquestioningly.
The more precise question is whether that contextual information genuinely supports the specific conclusion. A child's socioeconomic background, for example, does not establish their individual ability. An edited interview likewise may support a claim about what someone said without establishing their lasting character or private motive.
Three lenses, one recording
Media literacy asks who selected the material, supplied the introductory description and constructed the sequence. A particular edit may invite an interpretation, although this does not prove that the producer intended to mislead.
Information literacy distinguishes the visible performance from the broader conclusions drawn from it. A single recording may establish what was shown without providing a representative measure of ability, motive or behaviour across situations.
Digital literacy considers how excerpts, captions and audience commentary circulate. A short clip can lose information available in the full recording. Nevertheless, sharing a clip does not tell us whether the person forwarding it agrees with the implied interpretation.
What would reverse the judgement?
Before treating a mediated moment as evidence of a general trait or intention, consider what further observations could challenge that account. Would a longer sequence, an independent assessment or examples from different circumstances lead to another conclusion?
Evidence should help us decide between explanations, not simply decorate the one we formed first.
Key sources
Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labeling effects. Journal of Personality and Social Psychology, 44(1), 20–33.
Snyder, M., & Swann, W. B., Jr. (1978). Hypothesis-testing processes in social interaction. Journal of Personality and Social Psychology, 36(11), 1202–1212.
University of Texas at Austin, Ethics Unwrapped. Confirmation bias. Official educational resource.
Comments