Memory: Evaluation Flashcards
Multi-Store Model of Memory AO3
+ research support for 3 different stores: Baddeley found stm coded acoustically whereas ltm coded acoustically, Sperling found duration of SR 0.5 seconds, P+P found duration of stm 18 seconds
+ case studies: Clive Wearing had viral infection which meant he could form new STM not LTM, HM had operation to remove hippocampus and personality and intellect remained but couldn’t form new ltm,
- oversimplified: Shallice and Warrington (1970) studied KF who suffered brain damage, no problem with ltm, some aspects of stm impaired, difficulty dealing with verbal info but normal ability to process visual info
- doesn’t consider levels of processing theory: suggests long lasting memories created by type of processing rather than maintenance rehearsal, processed more deeply more memorable, Craik and Tulving (1975) found ps remembered most words from condition processing semantically
The Working Memory Model AO3
+ accounts for dual task performance: Baddeley + Hitch (1976) supports existence of CE, task 1 occupied CE, 2 either involved AL or CE and AL, 1 slower when 2 involved both CE and AL
+ studies of individuals with brain damage: KF short term forgetting of auditory info greater than visual, brain damage seemed restricted to PL and VSS not affected, LH performed better on spactial tasks than visual imagery tasks
- case studies: brain injury traumatic, changed behaviour, may have other difficulties eg paying attention, may underperform on certain tasks, unique individuals, can’t be generalised
- CE too vague and abstract: allocates resources and same as attention, critics feel notion of single CE is wrong, EVR performed well on tests requiring reasong suggesting CE intact, poor decision making suggests wasn’t entirely intact
Types of LTM AO3
+ neuroimaging evidence: Tulving et al (1994) had ps perform memory tasks in PET scanner, found episodic and semantic memories recalled from PFC, left involved in semantic, right involved in episodic
+ real life applications: allows psychologists to target certain kinds of memory to better peoples lives, Belleville et al (2006) demonstrated episodic could be improved in older people who had mild cognitve impairment, trained ps performed better on test of episodic after training than control
- difference is unclear: a lot of overlap, if remember first swimming lesson and know what swimming is and know how, which is it?
- problems with clinical evidence basis: learn lots of important info from studying brain injuries, lack of control of variables
Interference AO3
+ real world application in adverts: Danaher (2008) showed recall and recognition of ads message impaired when p exposed to 2 adverts for competing brands within week, one strategy to play ad multiple times in day rather than spreading it over week
+ temporary effect: Ceraso (1967) found if memory tested again 24 hrs later recognition showed spontaneous recovery recall remained same, interference occurs due to temporarirly not accessible
- artificial research: remembering lists of words don’t do irl, findings don’t relate to everyday uses of memory
- explains one type and some situations: does occur in everyday life but not often as memories need to be similar, disregards other types
Retrieval Failure AO3
+ context research support: Darley (1973) found ps who hid money while high more likely to find it when high again than not
+ context/state research support: Smith (1979) showed just thinking of room and environment where learnt as effective as being in room at improving memory
- cues don’t always work: info related to more than cues, learn about complex associations that less easily triggered by single cues, outshining hypothesis (cues effectiveness reduced by presence of better cues), Smith and Vela (2001) found context effects largely eliminated when learning meaningful material
- correlation: Nairne (2002) calss this ‘myth of the encoding-retrieval match’, encoding specificity model impossible to test because it’s circular, if stimulus leads to retrieval then it must have been encoded, if doesn’t then it can’t have been encoded but impossible to test for item that hasn’t been encoded
Misleading Information AO3
+ research support: Braun et al (2002) asked ps who visited Disneyland to evaluate ad material, misleading info about Bugs Bunny, asked to fill out questionnaires, 40% reported seeing or meeting Bugs Bunny
+ real world application: used to warn criminal justice system about problems with ewt, recent exoneration cases show incorrect ewt largest single factor contributing to conviction of innocent people, Innocence Project claim 72% of convictions overturned by DNA involved incorrect ewt
- research lacks external validity: Yuille and Cutshall (1986) interviewed 13 people who witnessed armed robbery 4 months after crime and included 2 misleading qs, witnesses provided accurate recall that matches initial reports
- individual differences: Anastasi and Rhodes (2006) found people in age groups 18-25 and 35-45 more accurate than 55-78, elderly have difficulty remembering source of info, more prone to misleading info
Anxiety AO3
+ C+H study of real crime: labs don’t create real levels of anxiety, Deffenbacher (2004) agrees but found lab studies demonstrate anxiety leads to reduced accuracy and real life studies associated with even greater loss
+ research support for both sides: Deffenbacher (1983) found 10/21 studies had results that linked higher anxiety to increased accuracy and 11/21 showed lower anxiety led to increased anxiety
- could be due to suprise not anxiety: Pickle (1998) had ps watch theif entering salon carrying scissors, handgun, wallet or raw chicken, identification least accurate in high suprise rather than high threat
- violence may affect accuracy: C+H concerned violent real life crime, other studies even real life didn’t, Halford and Milne (2005) found victims of violent crimes more accurate in recall
Cognitive Interview AO3
+ research support: meta analysis of 53 studies found increase of 34% in amount of correct info, volunteer witnesses in lab
+ useful when interviewing older witnesses: CI overcomes older adults overly cautious about reporting info, Mello and Fisher (1996) compared older and younger adults memory of filmed simulated crime using CI or standard, CI produced more info, strength of CI over SI greater for older
- amount of time and training: Kebbell and Wagstaff reported police officers suggest technique requires more time than available and prefer to use deliberate strategies aimed to limit ewt to minimum, CI requires special training and many forces haven’t been able to provide more than few hours
- effective in quantity rather than quality: procedure designed to enhance quantity without enhancing quality, Kohnken et al (1999) found 81% increase of correct info but also 61% increase of incorrect info when enhanced CI used compared to SI
Sensory Register Research AO3
+ Sperling lab study: high control, extraneous variables removed, true cause and effect, replicable
- Sperling lacks mundane realism: artificial task
STM Research AO3
+ P+P, Jacobs lab study: high control, extraneous variables removed, true cause and effect, replicable
- P+P, Jacobs, Baddeley artificial tasks: lacks mundane realism, more meaningful info recalled better, previous sequences may confuse future trials (proactive interference)
- Miller refuting research: Cowan (2001) reviewed studies on stm capacity and found limited to 4 chunks, Vogel (2001) looked at capacity for visual items said 4 chunks, Simon (1974) found shorter memory span for large chunks and longer for small
LTM Research AO3
+ Bahrick high mundane realism: task isn’t artificial, meaningful info
+ Baddeley lab study: high control, extraneous variables removed, true cause and effect, replicable
- Bahrick technology now affects it: can see former classmates on social media, remember them more, hard to replicate
- Baddeley artificial task: lacks mundane realism, meaningful info recalled better
Coding Research AO3
+ Baddeley lab study: high control, extraneous variables removed, true cause and effect, replicable
+ Baddeley makes ‘cognitive sense’: eg if remembering shopping list you’d repeat it out loud, if recall book you remember plot not every single word
- Baddeley artificial task: lasks mundane realism, meaningful info recalled better
Capacity Research AO3
+ size of chunk affects how many you can remember: Simon (1974) found shorter memory span for larger chunks eg eight word phases than smaller chunks eg one syllable words
+ Jacobs research method: lab experiment, high control, replicable
- Miller’s findings not been replicated: Cowan (2001) reviewed studies on capacity of stm and concluded limited to 4 chunks, research on capacity of stm for visual info also found 4 items limit (Vogel et al 2001), lower end of Millers range appropriate
- capacity of stm isn’t same for everyone: Jacobs found recall increased with age, 8 year olds remember average of 6.6 whereas 19 remember 8.6, maybe due to changes in brain capacity and/or development of strategies such as chunking
Duration Research AO3
+ Bahrick research method: natural experiment, recalling former classmates, high mundane realism
+ P+P research method: lab, high control, scientific
- artificial tasks: memorising consonant syllables doesn’t reflect everyday memory activities, do sometimes try to remember fairly meaningful things in groups of numbers of letters (phone numbers and post codes)
- P+P didn’t actually measure what it set out to: ps counting numbers in stm and may displace sullables to be remembered, Reiman (1974) used auditory tones instead of numbers so displacement wouldn’t occur and found duration of stm longer
