Ch.5-Working Memory: Flashcards
Working Memory
Ability to hold in mind and manipulate information. The mental workplace where the information we need to complete the process is stored and acted upon. Necessary for complex mental activities, short time periods.
Mental arithmetic: 25+38: hold the 2 numbers, use learned addition rules to compute solution, hold the sum (63) in working memory.
Other examples: remembering new password, remembering new name
Theory of Working Memory
Sensory memory–> short term memory–> long term memory
Distinguish between memory systems based on how long information is stored and the modality of the information.
Attending to information changes how it is represented and stored
Multiple Memory Systems?
Serial position effect: improved recall of words. At beginning of list: primacy effect
At end of the list: recency effect
Supports idea of different memory systems:
Primacy due to rehearsal (STM–> LTM)
Recency due to use of sensory memory or STM
Primacy and recency effects can be independently affected.
1st Stage: Sensory memory or Perception
Initial brief storage or sensory information. Information may or may not be attended to, some evidence encoding of sensory memory operates without deliberate attention (is automatic)
Separate stores may exist for each modality: Iconic (visual)
Echoic (audio)
Iconic Memory
Image persists even when no longer present (a mental event)
Sperling (1960): Iconic memory test. Bad at reporting whole display, extending display time did not help retention, asking for report of partial display did.
Stored whole display (iconic sensory memory): Lost information while reporting the first few letters
Erasing or Recoding the icon
Masking: Icon immediately followed by another stimulus
“mask” interrupts processing of icon and its information
information from icon is replaced from snesory memory by the “mask” and is thus no longer acted upon, does not make it to the next stage.
Interferes with remembering icon.
Erasing or Recoding the icon P.2
Icon recoded into a more meaningful representation after about 150-200ms. Higher level information about icon not available at first (eg: category of letter). Icon can only be recoded into another representation where meaning is available if the information is available for about 150-250ms. Recoding this information would allow it to be remembered for a longer period of time.
Echoic Memory
Suffix Effect: auditory word (but not beep or visual stimulus) erases list of “echoes” you are trying to remember, auditory masking
Sensory memories are modality specific, visual stimulus does not interfere with echoes
Second Stage: Short-Term Memory (STM)
Next level of storage and longer duration-usually under 1 or 2 min if rehearsal is not blocked.
Capacity of approximately 7+/- 2 items. Chunking smaller units into larger ones: NFLCBSFBIMTV–> NFL CBS FBI MTV. Increase amount of information we can process, overcome capacity limitation of STM.
Apply meaning to stimulus makes it easier to remember.
Conrad (1964)
Information mentally represented and held (coded). Conrad (1964) presented letters visually for participants to remember. When participants made errors, they confused letters based on sound rather than visual appearance (P confused for G)
Information presented visually, but mental representations formed in short term memory were acoustic not visual.
Baddeley(1966)
Gave participant a list of words (eg trip) to remember. Tested recall immediately or after delay. Immediate recall errors, studied words confused with ones that sound similar (trip vs trap).
Delayed recall errors: studied words confused with ones that have similar meaning (trip vs vacation)
Sachs (1967)
Participants listened to tape-recorded passages. Recognition test with sentences that were identical to those in the passage, sentences with change in meaning, sentences with change in grammatical structure. Participants decided if each sentence was old (in the passage) or new (not in passage)
What kinds of information would participants remember from the original passage: Falsely recognized sentences where the change was in grammatical structure but the meaning was the same
Sachs (1967) Example
Original: He sent a letter about it to Galileo, the great Italian scientist
Semantic Change: Galileo, the great Italian scientist, sent him a letter about it
Passive/active Change: A letter about it was sent to Galileo, the great Italian scientist
Formal: He sent Galileo, the great Italian scientist a letter about it.
Sachs (1967) Findings
Participants were best able to reject new sentences (not in the passage) if they had semantic changes. New sentences that still had the same meaning as the ones in the passage were falsely recognized. With more intervening material, more difficulty telling original sentence apart from changed sentences if changes were not semantic.
Original sentence stored with grammatical structure only until understanding is formed-meaning is stored and other forms of information are not.
Writing Good Sentences!
Reading a sentence requires short term memory storage. Hold parts of sentence in memory until they can be integrated with other parts of the sentence and a meaning developed. Nouns with similar meanings may be confused.
Which sentence creates a bigger memory load?
The senator who the president summoned arrived
The helicopter that the president summoned arrived
Short Retention Duration
Information from STM lost without rehearsal why?
Unrehearsed mental representation may decay. Alternately, additional information can displace old information, this is known as interference (masking)
Both decay and interference may be involved in loss of information from STM.
Brown-Peterson Task
The Brown-Peterson Task is a cognitive psychology experiment that demonstrates short-term memory (STM) decay and interference. In this task, participants are given a set of three consonants (e.g., BKG) to remember but are then immediately distracted by an interfering task—such as counting backward by threes—for a specific duration . If participants only count backward for 3 seconds, they still have a high chance of recalling the letters because the memory trace has not significantly weakened.
If they count for 18 seconds, recall drops dramatically. This supports the idea that memories in STM fade over time due to decay, meaning that without rehearsal, information is lost as time passes. The counting task prevents participants from rehearsing the letters, meaning they cannot refresh the memory in their phonological loop (a key part of working memory).
Making the letters meaningful – If the letters formed a real or familiar acronym (e.g., USA, IBM, CNN), recall would improve due to semantic encoding.
Using similar-sounding letters – Letters like B, P, V, D, T are easily confused due to phonetic similarity, increasing recall difficulty.
Searching our STM
Sternberg-memory set consisting of 1-7 items (BKFQAZW) encoded, decide if probe was in memory set (K). Position of probe in set (K vs W) did not seem to influence time taken to answer. Serial exhaustive search as the way we search STM. Search process may be so rapid that it is difficult to stop even after we find probe.
Rebranding Memory: Working Memory
WM is a reconceptualization of STM. Conceptualized by Baddeley and Hitch (1974). Stores pieces of information for a short period of time, but there is more to WM.
Working memory involves active processing of information, making information available for other cognitive processes, integrating various pieces of information, transferring information to Long-term memory and retrieving relevant information from LTM.
Working Memory: Workflow management
WM is more like a space that displays information for several components that act on it. Limited capacity temporary storage system that consists of several components: Central executive, phonological loop, visuospatial sketchpad, episodic buffer
Central Executive
Directs the flow of information and allocates resources to cognitive tasks: Attentional process, coordinates incoming external information with retrieval of relevant information from LTM, choosing which information will be operated on when and how, allocating resources to cognitive tasks.
Processes of central executive equated with conscious awareness (Baddeley 1993)
Temporary Storage/maintenace of info: Phonological Loop
Subvocal rehearsal to maintain verbal material.
Important in creating mental representations of sounds.
Learning new words
Temporary Storage/Maintenance of info: Visuospatial sketchpad
Visualization to maintain visual material.
Creation and use of mental images.
Temporary Storage/maintenance of info: Episodic buffer
Integrates information from phonologicall loop and visuospatial sketchpad.
Controlled by central executive.
Facilitates transfer of information into and from LTM.
