You are viewing a free preview of this lesson.
Subscribe to unlock all 9 lessons in this course and every other course on LearningBro.
Neville Moray's 1959 study is the classic study for the cognitive-area theme of attention, and it is one of the foundational experiments of the modern psychology of attention. Its subject is a puzzle everyone has lived: at a noisy party, surrounded by many simultaneous conversations, we can select one voice to follow and screen the rest out — yet if someone across the room says our own name, it will often "break through" and grab our attention even though we were not listening to that conversation at all. This everyday miracle of selection-with-occasional-leakage is what Moray set out to pin down experimentally, and his findings became a central piece of evidence in the great mid-century debate about how the mind filters the flood of information arriving at the senses.
This lesson tells the study in the OCR "tell the story" format: the background — including the filter theory of Donald Broadbent that Moray was testing — the aim, the method (design, sample and the step-by-step procedure of each experiment), the results with their real figures, Moray's conclusions, and a full evaluation of its method, data, ethics, validity and reliability. It closes by linking the study to its key theme, its area, the relevant perspectives and the debates it fuels. As the classic partner to Simons & Chabris (1999), Moray anchors the attention theme, so knowing his procedure and findings precisely — and knowing how attention research moved on from him — is essential for Component 02.
| This lesson covers | OCR H567 Component 02 element | AO focus |
|---|---|---|
| Background: selective attention, the cocktail-party problem, Broadbent's filter theory | Section A — Cognitive; theme: attention (classic) | AO1 knowledge |
| Method: dichotic listening and the shadowing technique; three sub-experiments; sample | Section A — Core study (Moray) | AO1; AO2 |
| Results (near-zero recognition of the rejected message; the own-name effect; instructions) | Section A — Core study | AO1 |
| Conclusions (attention as a limited-capacity filter; the block is not wholly impermeable) | Section A — Core study | AO1; AO3 |
| Evaluation: method, data type, ethics, validity, reliability, sampling, ethnocentrism | Section A; Section B debates | AO3 |
| Links to theme, area (Cognitive), perspective and debates | Section B — Areas, perspectives, debates | AO1; AO3 |
The specification is referenced descriptively; consult the official OCR H567 specification document for its exact published wording. This lesson develops AO1 (the aim, procedure, results and conclusions), AO2 (applying the filter account of attention to novel listening scenarios) and AO3 (evaluating the study's validity, reliability, ethics and generalisability).
By the late 1950s the psychology of attention was being transformed by two forces: the practical problems of the wartime and post-war years, and a new theoretical vocabulary borrowed from communications engineering. During the Second World War, researchers had grappled with real questions about how air-traffic controllers and radio operators could follow one voice among many competing signals — problems in which the limits of human attention had life-or-death consequences. Out of this work came the idea that the human mind, like a communication channel, has a limited capacity: it cannot fully process everything arriving at the senses at once, and so must select.
The most influential expression of this idea was Donald Broadbent's filter theory of attention. Broadbent proposed that information from the senses passes first into a brief sensory store, and then through a selective filter that lets only one input channel through for full processing while the rest are blocked. Crucially, in Broadbent's original account the filter selects on the basis of physical characteristics of the input — such as which ear a message arrives in, or the pitch of a voice — and it does so early, before the meaning of the rejected messages is analysed. On this view, an unattended message is filtered out on physical grounds and its meaning is never processed at all. This is the "early selection" model, and it makes a strong, testable prediction: we should retain essentially nothing of the content of a message we are not attending to.
The classic tool for testing such predictions was the dichotic listening task, in which a participant wears headphones and hears a different message in each ear simultaneously. To ensure they attend to only one of the two messages, they are asked to shadow it — to repeat it aloud, word for word, as they hear it. Shadowing is a demanding task that occupies attention almost completely, so it operationalises "attending to one channel" as a measurable behaviour, and it lets the experimenter ask precisely what, if anything, gets through from the other, unattended ear.
Moray set out to test Broadbent's filter theory and, in particular, to probe its limits. He suspected that the block on the unattended channel might not be quite as total as strict early-selection implied — that some especially significant material might slip through. The everyday observation that motivated him is the famous "cocktail-party phenomenon" (a phrase from the engineer Colin Cherry, whose earlier shadowing experiments Moray was building on): the fact that we can follow one conversation in a noisy room, yet our own name spoken elsewhere can capture our attention. If our own name can break through, then the filter cannot be blocking the unattended channel purely on physical grounds without any regard to meaning — because recognising your name requires that its meaning be processed. Understanding this theoretical stake is what makes Moray's results significant rather than a mere curiosity: they bear directly on where and how the attentional filter operates.
The overarching aim was to investigate the nature of selective attention in hearing — specifically, how much of a rejected (unattended) auditory message is processed, and under what conditions material from the rejected channel can nonetheless break through into awareness. Moray pursued this through three linked experiments, each with its own aim:
All three were laboratory experiments using the dichotic listening task with the shadowing technique. Participants wore headphones and heard two spoken messages simultaneously, one in each ear; they shadowed one of the two (repeating it aloud) so as to attend to it, while a different message played in the other, rejected ear. Messages were pre-recorded on tape so that the procedure was tightly standardised and identical for every participant. Across the three experiments the design varied — Experiments 2 and 3 involved comparisons that used a repeated-measures logic (the same participants experienced the relevant conditions) — but the core apparatus and shadowing method were common throughout.
The participants were undergraduate students and research workers of both sexes, tested at the University of Oxford. The number varied by experiment (a small sample by modern standards — of the order of a dozen or so per experiment, with a larger pool across the whole study), consistent with the intensive, individually-run nature of dichotic-listening research. This is an opportunity sample of academically able young adults, a point that matters for the evaluation of generalisability below.
Procedure. Participants shadowed a prose passage presented to one ear. Simultaneously, in the other (rejected) ear, they heard a short list of simple words that was repeated many times — on the order of 35 repetitions — while they were shadowing. Immediately after the shadowing task ended, participants were given a recognition test: they were presented with a list of words and asked which they had heard, the list containing some of the words from the rejected message, some from the shadowed message, and some entirely new words that had appeared in neither. The measure was how many of the rejected-ear words participants could recognise.
Procedure. Participants shadowed a prose message in one ear while a different message played in the rejected ear. Embedded in the messages were instructions — for example, "change to the other ear", or instructions addressed to the listener. Some of these instructions in the rejected ear were prefaced by the participant's own name ("John Smith, change to your other ear…"), and some were not (an impersonal instruction with no name). The comparison of interest was whether instructions in the rejected ear were more likely to be heard and obeyed (or noticed) when they were preceded by the listener's own name than when they were impersonal. Because each listener could receive both named and un-named instructions, the affective-cue comparison used a repeated-measures logic.
Procedure. This experiment tested whether directing participants in advance to listen for particular content could break through the block. Participants shadowed one message; both the shadowed and the rejected messages contained digits (numbers) inserted into the prose. In one condition participants were instructed beforehand that they should try to remember numbers (or were led to expect that numbers were important); in another they received no such instruction. The measure was whether numbers embedded in the rejected message were remembered better when participants had been set to attend to numbers than when they had not — that is, whether a deliberate "set" for a category of material could pull that material through from the unattended channel.
The result strongly supported Broadbent's early-selection prediction: participants recognised almost none of the words from the rejected ear, even though those words had been repeated around 35 times while they shadowed. Recognition of the rejected-ear words was at essentially chance level — no better than for the entirely new words that had never been presented — whereas words from the shadowed message were recognised well. In short, the sheer repetition of material in the unattended ear left virtually no trace in memory. This is powerful evidence that, under normal conditions, the content of an unattended auditory channel is blocked and not stored.
Here the block proved permeable to personally significant material. Instructions in the rejected ear were noticed and acted upon far more often when they were preceded by the participant's own name than when they were impersonal. In Moray's data, the participant's own name in the rejected ear was detected on roughly a third of the occasions it was presented (in the region of 20 out of 39 presentations across the sample), whereas impersonal instructions in the rejected ear were very rarely noticed. The contrast is the study's most celebrated finding: the same channel that leaves ordinary words untraced (Experiment 1) lets one's own name through a substantial fraction of the time. This is the experimental capture of the cocktail-party phenomenon.
In contrast to the own-name effect, a deliberate set to listen for numbers did not significantly improve memory for numbers embedded in the rejected message. Participants who had been told numbers were important were no better at remembering the rejected-ear numbers than those who had not — the numbers in the unattended channel were not selectively pulled through by the instruction. (Any effect of the instruction operated on the shadowed message, not the rejected one.)
Taken together, the three results paint a coherent picture: the unattended channel is largely blocked (Exp 1), a simple advance instruction to attend to a category of material does not breach the block for the rejected message (Exp 3), but personally significant material — one's own name — does breach it (Exp 2).
Moray drew several connected conclusions about the nature of selective attention.
First, and consistent with Broadbent, little information from a rejected auditory message is processed or retained: the near-chance recognition in Experiment 1 shows that under normal conditions the unattended channel is effectively blocked, so almost none of its content reaches memory. Attention behaves as a limited-capacity selective filter that admits one channel and screens out the others — supporting the general filter-theory framework.
Second, and more subtly, the block is not completely impermeable. The own-name effect (Experiment 2) shows that some material from the rejected channel does get through — specifically, material of personal importance or affective significance. Because recognising one's own name requires that its meaning be analysed, this finding is difficult to reconcile with a strictly early filter that blocks the unattended channel on purely physical grounds before any meaning is processed. It suggests instead that the unattended message is analysed for meaning at least enough for highly significant items to be detected — a point that would later drive theories placing the filter later (or making it attenuate rather than wholly block the rejected channel) than Broadbent's original model.
Third, the breakthrough is selective, not general. The contrast between Experiment 2 and Experiment 3 is important: it is not the case that any instruction or any category of material can be pulled through the block. A neutral advance set for numbers did not breach it (Exp 3); only personally significant material did (Exp 2). This implies that what determines breakthrough is the subjective importance of the material to the listener, not merely being told to look out for it.
Subscribe to continue reading
Get full access to this lesson and all 9 lessons in this course.