I.  Sensory Memory

    A. In part, our memories are records of percepts

        Briefest memory stores last for only a fraction of a second

        An integral part of the process of perceiving

    B. Memory based on a particular sensory modality

         Both vision and hearing appear to have a later but temporary storage stage

         Also have long-term memory for sights and sounds

    C. Iconic and echoic memory represent the earliest stages of sensory memory

         Similar systems probably exist for the other senses

II. Visual memory or Iconic memory

    A. The persistence of vision

        How much visual info. can be gathered in one glimpse?

        Span of apprehension

        Tachistoscopic presentation

        Averbach (1963)--number of dots that could be accurately reported depended on the exposure duration    

    B. Partial Report Technique

        Sperling (1960)—presented Ss with a stimulus comprising three rows of four letters.  When shown these for 50 msec, followed by a blank white field, Ss were able to report only about four or five of the twelve letters. 

        Were these the only letters that had been seen, or could the S have seen them all, but forgotten some of them in the time it takes to report five letters?  

        S cued to report only one line of letters

        How should Sperling’s results be interpreted?  

       Sperling described iconic memory  as precategorical

        B.  The nature of the trace

         Once the item is scanned, need to rehearse or the information will be lost. 

        When target letters are followed by a bright flash of light, performance is substantially impaired, as if the flash of light were wiping out the icon.  

        Auditory confusion errors--tendency for errors to be similar to the correct item in sound rather than in visual characteristics.  Suggests speech-based coding of visually presented letters. 

        C.  Ecological function of iconic memory  

III.  Auditory Memory or Echoic Memory

    A. The persistence of audition 

       Efron (1970)--presented a series of brief tones and asked Ss to adjust the onset time of a light to coincide exactly with the end of the tone.  Regardless of whether the tone actually lasted for 30 ms or 100 ms, its subjective apparent duration was approximately 130 ms.  Resembles the case for iconic memory.

         What is the rate at which the trace decays? 

Glucksberg and Cowan (1970)—required Ss to shadow prose presented to one ear.  At the same time, they presented prose to the other ear which the S was instructed to ignore.  Green light appeared intermittently, signaling S to say whether a digit had just appeared to the unattended ear.

         Unattended items appear to be lost within about 5 seconds.

B. Partial Report Technique 

Darwin, Turvey, and Crowder (1972)

  Three lists of three digits.  Played 3 lists simultaneously from 3 different locations.  

         Clear advantage for partial report 

Capacity is not as large as with the visual modality, but it lasts longer 

Advantage of partial report almost disappears after 4 seconds

       Echoic memory appeared similar to iconic memory with just two minor differences: (1) echoic memory lasts longer (approx. 4s compared to approx. 1 s for iconic), but (2) does not preserve as much information (about 4 or 5 items compared to 8 or 9 for iconic).  

In-depth analysis of serial recall accuracy for partial report .  Strong recency effect and moderate primacy effect. 

C. Modality and Suffix Effects 

Modality effects—In general, Ss are presented a series of items (digits 1-9), and are asked to recall them in order (serial recall).  

Results--Auditory and visual presentation produce almost the same result.  Exception:  Large advantage for the last one or two items in the list with auditory presentation (recency effect) compared to visual presentation.

Conrad & Hull (1968)--Ss read items aloud or silently  

Big advantage for the last item (80% correct recall vs. about 20% correct recall) for auditory presentation. 

Suffix effect—applies only when the presentation modality of the to-be-remembered items is auditory.  

In the suffix condition, an extra item follows the last to-be-remembered item.  

Subject is told to ignore the suffix or to treat it merely as a sign to begin recall. 

Suffix dramatically reduces recall of the final item. 

Crowder and Morton (1969)—non-speech sound (buzzer) did not disrupt performance.  

Meaning of the suffix was non important; any speech sound will impair performance. 

Interpretation—brief precategorical acoustic store.

Trans A

Trans B

Suffix