What Do You Hear?
Updated: Aug 2
Auditory Perception, Expectation, and the Formation of Memory
Two people can listen to the same sound and sincerely report hearing different things.
This can occur when people listen to a low-quality recording, but it is not limited to recorded speech. It can also occur during ordinary conversation, emergency events, workplace incidents, interviews, transportation events, public-safety encounters, and other rapidly unfolding situations. People standing within feet of one another may hear the same speaker, the same command, or the same sequence of sounds and later provide substantially different accounts of what occurred.
These differences do not necessarily indicate dishonesty, carelessness, or a failure to cooperate. They may reflect the way human perception and memory actually function.
Before continuing, view the Institute’s brief demonstration, “What Do You Hear?” The video illustrates how the same ambiguous auditory signal may support different interpretations and how expectation can change what a listener believes the sound contains.
Hearing Is an Interpretive Process
Sound enters the auditory system as patterns of acoustic energy. The listener must organize that information into meaningful objects, voices, words, and events. In clear and predictable conditions, this process usually feels immediate and effortless. We experience the result as though we simply heard what was objectively present.
The underlying process, however, is constructive.
The brain interprets incoming sound by combining the available acoustic information with context, language knowledge, expectations, attention, prior experience, and information from other senses. Classic research has demonstrated that visual information can alter what a person believes was spoken. In the well-known McGurk effect, conflicting visual and auditory speech information can produce a perceived sound that differs from the acoustic signal itself.
Research on phonemic restoration provides another example. When part of a spoken word or sentence is replaced by another sound, listeners may perceive the missing speech sound as though it were actually present. The perceptual system uses linguistic context to create a coherent interpretation from incomplete information.
These processes are normally helpful. Everyday speech is frequently incomplete or degraded by distance, competing voices, reverberation, machinery, traffic, wind, radio transmission, protective equipment, poor acoustics, and background noise. Human communication would be extraordinarily difficult if listeners could interpret only perfectly transmitted speech.
The same processes that make speech understandable, however, can also support more than one interpretation when the signal is ambiguous.
The Role of Expectation
The video accompanying this article presents an ambiguous sound and demonstrates how a listener’s interpretation may change after being given an expectation about what the sound might contain.
Before a suggestion is introduced, the listener may hear an indistinct sound, an uncertain word, or one of several possible phrases. Once a particular interpretation is supplied, the same acoustic signal may begin to sound clearer. The listener may then experience the suggested phrase as though it had always been obvious.
The recording has not changed. The listener’s perceptual framework has.
This is an important distinction. A person may sincerely report what the person experienced hearing, but that experience is not necessarily a direct reproduction of the underlying acoustic signal. It is the product of the signal interacting with the listener’s expectations and interpretive processes.
Expectation may arise from many sources:
knowledge of what is supposed to happen;
familiarity with a speaker or situation;
prior conversation;
visible movements or gestures;
the reactions of other people;
occupational experience;
emotional significance;
an investigator’s question;
a written transcript; or
another person’s interpretation of the sound.
A listener who is told what to expect may have difficulty returning to the original state of uncertainty. The suggested interpretation can become perceptually dominant.
Recorded Speech and Real-Time Speech
Ambiguous recorded speech makes this process easy to demonstrate because the identical signal can be replayed. Several people can hear the same file repeatedly, and the file itself remains unchanged. Differences among listeners therefore cannot be explained by changes in the recording.
Real-world auditory events are more complex.
During an unfolding event, each person occupies a different physical and cognitive position. Even people standing close together may not receive identical information. One person may be facing the speaker while another is turned away. One may see the speaker’s face, while another has an obstructed view. One may be attending to the conversation, while another is watching a vehicle, a weapon, an instrument panel, a patient, or another person.
Small differences in location may also affect the acoustic signal. Distance, room geometry, reflective surfaces, competing sounds, head orientation, and the direction in which a person is speaking can influence clarity. Simultaneous speech may mask part of a sentence for one listener but not another.
The listeners may also bring different expectations to the event. A person who anticipates a warning may interpret an unclear utterance as a warning. Another person, expecting reassurance, may interpret the same utterance differently. A trained professional may organize sounds according to occupational knowledge that another witness does not possess.
Consequently, people who were present at the same time and in nearly the same place may not have experienced the event in precisely the same way.
Perception Becomes Part of Memory
Memory does not begin after an event ends. Its formation begins while the event is being perceived.
What is encoded into memory is not a complete audiovisual recording of the environment. It is a selective and interpreted representation of what received attention and what the person understood at the time.
If a listener initially interprets an ambiguous statement one way, that interpretation may become part of the memory of the event. Later, the person may remember not only that a sound occurred, but that a particular phrase was spoken. The distinction between the original acoustic information and the meaning assigned to it may no longer be available to conscious awareness.
Language encountered after the event can also influence subsequent memory. Research has long shown that the wording used to ask about an event can affect later estimates and recollections.
This does not mean that every memory is inaccurate or that witness evidence should be disregarded. A more useful framework is to recognize that memory evidence can be preserved or contaminated. Early, independently obtained accounts may provide valuable information, while suggestion, repeated exposure to interpretations, and discussion among witnesses can alter later reports.
Why Witnesses Remember the Same Event Differently
Consider several witnesses who simultaneously experience a brief and consequential event. They may later disagree about:
the exact words that were spoken;
whether a command or warning was given;
which person spoke first;
the number or sequence of sounds;
whether a voice sounded angry, frightened, calm, or urgent;
whether two sounds occurred separately or nearly together; or
what a particular sound represented.
These differences may emerge at several stages.
First, the witnesses may have received different sensory information because of their locations, orientations, and environmental conditions.
Second, they may have allocated attention differently. Human attention is selective, particularly during complex or rapidly changing events.
Third, they may have interpreted ambiguous information differently based on prior knowledge and expectations.
Fourth, they may have encoded different aspects of the event into memory.
Finally, later conversations, questions, media exposure, transcripts, recordings, or statements from other witnesses may influence how the event is remembered and described.
Research on memory conformity demonstrates that information provided by co-witnesses can affect later reports, particularly when a person is uncertain about what was originally perceived.
Two substantially different witness accounts can therefore originate from the same event without either witness intentionally providing false information.
Confidence and Perceptual Clarity
A particularly important feature of auditory interpretation is that subjective clarity may increase after contextual information is supplied.
Once a listener hears a suggested phrase in an ambiguous recording, the phrase may seem unmistakable. The listener may become highly confident because the perceptual experience itself feels clear.
Confidence in that later perception, however, does not establish that the phrase was independently perceived before the suggestion was introduced. It may demonstrate how effectively the supplied context organized the ambiguous signal.
For this reason, investigators, analysts, attorneys, researchers, and decision-makers should distinguish among:
what the person initially heard;
what the person initially believed the sound meant;
what information was introduced afterward; and
what the person hears or remembers after receiving that information.
These are related but not interchangeable questions.
Implications for Investigative Interviewing
The way auditory memories are elicited can affect their evidentiary value. Interviewers should seek the witness’s original account before introducing interpretations from recordings, transcripts, other witnesses, or investigators.
Useful practices include allowing uninterrupted narrative recall, using open-ended prompts, documenting the witness’s own words, and preserving expressions of uncertainty. A witness who says, “I heard something that may have been a warning, but I could not make out the words,” is providing meaningful information. That uncertainty should not be treated as a gap that must be filled.
Questions that contain a proposed phrase can unintentionally create a new listening framework. Asking, “Did you hear him say, ‘Drop the weapon’?” is cognitively different from asking, “What, if anything, did you hear the person say?”
The Cognitive Interview was developed to improve retrieval while reducing unnecessary interviewer influence. Research comparing enhanced Cognitive Interview procedures with conventional approaches has supported the value of structured, witness-centered retrieval methods.
When multiple witnesses are involved, obtaining separate accounts before they discuss the event can help preserve independently formed memories. Investigators should also document whether a witness has listened to recordings, read transcripts, watched media coverage, or discussed the event with others.
Implications for Recorded-Audio Analysis
Ambiguous recordings require similar caution.
A proposed transcript should not be treated as though it were the acoustic evidence itself. Once listeners read a suggested phrase, they may begin to hear the recording through that linguistic framework. Competing interpretations should be considered, and listeners should be informed when the signal does not support a definitive conclusion.
Repeated listening does not always resolve ambiguity. In some circumstances, repetition may strengthen a preferred interpretation without adding new acoustic information.
Whenever possible, the original recording should be preserved and evaluated using appropriate technical methods. Analysts should distinguish between:
the measurable properties of the recording;
speech or sound that is clearly intelligible;
speech that is partially intelligible;
competing plausible interpretations; and
content that cannot be reliably determined.
The goal should not be to force every sound into a word or phrase. The goal should be to identify what the evidence supports and to preserve the boundary between observation and interpretation.
Understanding Variability Without Discrediting the Witness
Differences among witnesses are sometimes treated as proof that one account must be truthful and another false. Human perception and memory do not support such a simple conclusion.
A witness can be sincere, attentive, and confident while still providing an account shaped by limited sensory information, expectations, selective attention, and later influences. Another witness may remember a different aspect of the same event with equal sincerity.
Understanding these processes does not require abandoning witness evidence. It requires collecting and evaluating that evidence with greater care.
Variability among accounts may itself provide useful information. It may identify which portions of an event were perceptually clear, which were ambiguous, which witnesses had different vantage points, and where later information may have influenced recall.
Conclusion
The question “What do you hear?” appears simple, but the answer is produced by more than our ears.
The Veritas Method™ examines how perception, expectation, attention, and memory influence what people believe they heard and how they later describe an event. This topic is addressed directly in the course as part of understanding why sincere witnesses may provide different accounts of the same experience and how interviewers can obtain information without unnecessarily shaping recall.
Auditory perception reflects an interaction among acoustic information, context, attention, expectation, experience, and interpretation. That perceived meaning can then become part of memory. When an event is ambiguous, rapidly unfolding, or emotionally significant, different people may encode and later recall substantially different versions of what occurred.
Those differences are not necessarily evidence of deception. They are often evidence of human perception.
Recognizing this distinction is essential in investigative interviewing, forensic audio analysis, safety investigations, legal proceedings, organizational reviews, and every setting in which people are asked to describe what they heard and remember.
Comments