Memory & Intelligence Flashcards
What is known as the “mental glue” that binds one moment to the next?
Memory
What’s memory?
The process of using information that was obtained in the past in order to generate some cognitive function in the present
Memory is any cognitive process that involves what 3 fundamental components (memory processing stages)?
- Encoding
- Storage
- Retrieval
What’s encoding?
- The initial processing of information by the nervous system -> process of inputting info into memory
- Creating separate memory traces to represent experiences
- This may be in the form of a short-term transduction of a physical stimulus into a neural code or a structural change in the brain that encodes a fact or event about the world
- This process is the necessary predecessor of storage and retrieval -> if something isn’t encoded in the nervous system, it can never be remembered
- Here, a memory trace is formed as a hippocampal-cortical activity pattern
- Encoding is when information from the short term is transformed into the long term
What’s storage?
- The retention of information in the nervous system beyond initial processing (retention of encoded information or encoded memory traces)
- Encoded information by the nervous system remains encoded in some form for a longer duration than immediate processing
- Here, via consolidation, a memory is transformed into a stable cortical pattern
What’s retrieval?
- The access and use of stored information by the nervous system for some cognitive purpose
- A memory is recovered when a cue activates part of a stored memory trace
- That memory trace then enters conscious experience
- Here, part of a memory trace is activated by a cue that triggers pattern completion
- When LTM retrieves info from STM or WM
What happens if any of the links in the chain of memory (encoding, storage, retrieval) is broken?
Memory can’t function
What’s analogous to human memory?
Computer memory
Why are human memory and computer memory analogous?
- Both human and computer memory require the same basic elements to function: a means of encoding, storing and retrieving information
- Like human memory, a failure at any stage of this process will lead to a memory “loss”
What is one of the most important recent advances with regard to the emergence of a highly computer-connected human society?
The development of powerful search engines that enable successful (human-driven) retrieval of information from the “memory” of the internet
Describe how encoding, storage and retrieval can be used when describing to a friend that you saw a mutual acquaintance at an event last week
- This requires initial encoding at the event, where your brain processed the light and images entering via your eyes’ retinas. This allows you to see your surroundings and mutual acquaintance
- This visual processing by the brain leads to a cascade of storage activities involving multiple cortical and subcortical brain regions that can store the information for both short and longer-term durations
- Finally, retrieval of that stored memory allows you to describe the event, activating your language and motor areas in order to produce the speech you use to tell your friend what you saw
What are 2 basic dimensions of memory that we can measure with behavioral research?
- We can measure how much information a memory system can hold (its capacity)
- We can measure how long information remains in memory (its duration)
- Considering and contrasting these 2 dimensions across different conditions has been sufficient for investigators to draw firm conclusions that there are different kinds of memory that exist in the human brain
What’s the memory’s capacity?
- A measure of how much information a memory system can hold
- Some types of memory appear to have much higher capacity than others
- Ex: there’s a very limited amount of numbers or letters that you can hear & repeat verbatim -> a task that requires short-term memory (no such limit for the amount of more general info you can retain in long term memory)
What’s the memory’s duration?
- A measure of how long information can be held in memory
- Some types of information are very quickly lost from memory while other information may be retained for a lifetime
What’s one of the most important and influential concepts in the history of cognitive psychology?
The idea that memory isn’t one kind of function but that instead there are different types of memory
Who was the first person to formally articulate the idea that there may be 2 kinds of memory stores and what were those 2 kinds of memory stores?
- William James
- One for information related to the current task or environment
- One for longer-term storage
- However, these were based on casual observation rather than experimental research
What was the first substantial theoretical model of memory that attempted to account for experimental data?
Atkinson & Shiffrin’s (1968) multi-store/modal model of memory
Describe Atkinson & Shiffrin’s (1968) multi-store/modal model of memory
- An influential model of human memory that posited 3 distinct memory stores: sensory, short-term, long-term
- All of these different memory stores work in conjunction with one another during everyday cognition
- This model proposed that each of these memory stores have their own duration and capacity
- The modal model was created in a new era of computer science
- It likens our processing of information to encoding and storing information into a computer
- AKA information processing model
What’s the first stage in the modal model of memory?
Sensory memory
What’s sensory memory?
- 1st and “temporary” stage in the modal model of memory
- Holds information before it can be processed
- Has a high capacity and short duration
- It briefly stores the info just encoded by the sensory organs
- The purpose of this form of memory is simply to hold the information in place before it can be selected, via attention, for further processing
- Only a fraction of the information stored in sensory memory ends up being selected by attention and passed along to short-term memory (STM)
- Under some conditions, sensory memory may actually be directly observed in action (ex: persistence of vision phenomena)
- Duration ~1 second
- Important for when we have to make quick decisions about what’s in our environment
What’s short-term memory (STM)?
- 2nd stage of the modal model of memory
- Serves to hold processed information for rehearsal or to produce a behavior
- Has a much smaller capacity than sensory memory (magical number seven plus or minus two”) but a considerably longer duration (approx. 15-30 secs)
- Unlike sensory memory, STM is capable of producing a behavioral output, such as repeating a phone number someone has just told you or responding to a recall task
- STM has the capacity for reactivating information stored within it through a process called maintenance rehearsal
- Involves the prefrontal cortex
- Attended information moves from sensory to short term memory
What’s maintenance rehearsal?
- The mental repetition of information in short-term memory that allows information to be regenerated in order to prolong its duration
- This process essentially reactivates the initial encoding
- Once information is rehearsed, it can restart the clock on the duration of the memory
- A technique for encoding information in long-term memory (LTM)
- Without any distractions, people are able to engage in maintenance rehearsal
- If you have absolutely nothing else to do, information can theoretically persist in STM for as long as you can keep rehearsing. However, in the real world, this isn’t possible
- Some research has found that such repetition can lead to a modest bump in retention in LTM
- However, simple repetition isn’t as effective as techniques in which we go beyond the simple repetition of the info in STM and consider the meaning of the info
What’s long-term memory (LTM)?
- The final stage of the modal model of memory
- Serves as cold? storage of information for retrieval into short-term memory
- Because of the nature of how LTM is encoded, there’s no agreed-upon method for measuring its capacity, but it’s clear that only a small fraction of STM is encoded as LTM
- LTM comprises anything that’s remembered from beyond roughly 15-30 secs (duration of STM) up to a lifetime
- We know that at least some memories can last a lifetime but that many decay over time as well
- The duration and capacity of LTM have no quantified limit
- Theoretically LTM is infinite -> lifetime memory and no known capacity limit
- The lack of duration and capacity limits of LTM doesn’t mean that all of the information in STM makes its way to and is retained for a long period in LTM -> most of the info we encounter in our lives doesn’t seem to be stored in LTM, at least not for very long
- According to the modal model, STM items gradually transfer to LTM
- Unlike STM, LTM doesn’t seem to depend on continuous activation of the original activation used to encode the stored info
There’s evidence that after the initial transduction of sensory information, it’s retained within our nervous system for how long?
A brief period of around 250 ms
What’s the persistence of vision phenomenon?
- The retention of an image of an object or event for a brief period of time after it’s no longer present
- Here you can directly see information that entered your eyes moments ago
- Ex: when an object moves really fast and it looks spread out over space
What’s an early example of the persistence of vision?
- A 19th Century toy called the thaumatrope
- A disk with different drawings on each side and when the disk was spun quickly, it would lead to the illusion of seeing both images at the same time
Describe Sperling’s (1960) experiments on sensory memory
- He conducted a series of well-known experiments on sensory memory in which participants were presented with stimuli consisting of 3 rows of 4 letters (12 letters total) -> the display flashed for .05 seconds
- There was a whole-report condition and a partial-report condition
Describe the whole-report condition of Sperling’s experiments on sensory memory
- In the whole-report condition, participants were flashed a grid of letters for a duration of .05 ms and were asked to report as many of the letters from the whole display as they could after the beep sound
- He found that participants typically reported the letters from one of the rows, suggesting that whichever row they happened to be paying attention to was available to report
- This ability to report the letters, a behavioral output, depends on those items making their way into STM
- This indicated that approx. 25% of the overall grid could be retained in STM (~4 items)
Describe the partial-report condition of Sperling’s experiments on sensory memory
- In the partial-report condition, participants were given instructions to report only one specific row of letters in the grid based on the presentation of an auditory tone whose pitch indicated which row participants had to report (within .01 - 1 second, a high/medium/low tone signals which row to report)
- On critical trials, they were only provided with this tone after the letter array had flashed -> meaning that by the time participants were told which row to report, the stimulus was already gone
- In this condition, participants were shown a grid of letters for .05 seconds
- Participants had to rely on their memory of the grid to report a specific row
Describe the findings for the partial-report condition of Sperling’s experiments on sensory memory
- He found that if the tone was presented very soon after the letter grid had disappeared (ex: 100 ms), participants were often able to report 75-80% of the letters in the row that was indicated
- This suggests participants could actually attend to that line of letters and call it into STM, even though the physical stimulus was no longer present
- He concluded that much more than 25% of the letter grid was still present in sensory memory and that, during that time, the participants could decide which of the rows to pay attention to in order to bring them into STM and report them
- He also found that as he increased the delay between when the grid disappeared and when the tone was presented, the capability to accurately report a specific row of letters decreased
- By the time the delay was 1 second, participants were no longer able to use the tone usefully and they only reported whichever row they happened to have been attending to
- Here the grid was no longer available in sensory memory and performance was equivalent to the whole-report condition (accuracy of approx. 25% of the grid)
- This study tells us that the sensory memory capacity is quite large, people could remember or recall a lot of info from sensory memory but only for a short period of time
What did Sperling’s experiments on sensory memory suggest about the duration of sensory memory?
That the duration of sensory memory is approx. 1 second
Describe the process of forgetting with sensory memories
Like other kind of memory, “forgetting” takes place over time, with better memory immediately after stimulus presentation and gradually fading away until nothing is left after a second
What’s iconic memory?
- The visual form of sensory memory in which much of the visual input can be stored for a short period
- Sperling referred to this high-capacity/short-duration form of visual memory as iconic memory, based on the fact that the memory is something like a photographic image or icon
What’s echoic memory?
- An auditory form of sensory memory in which much of the auditory input can be stored
- Sound-byte held for ~ 3 seconds
- Important for distinguishing between when 1 person is speaking and when another is speaking
STM only holds information that has been selected by what for processing?
- Attention
- Attended information moves from sensory to short term memory
How many items can STM hold in the auditory domain?
- It’s generally agreed that it can hold an average of 7 items
- Limited capacity: magical number 7 plus or minus two”
- Ex: remembering a list of names or numbers
How many items can STM hold in the visual domain?
It’s agreed (not unanimously) that it can hold around 4 items
How did Miller (1956) describe the surprisingly stable capacity of auditory STM?
- He presented participants with lists of letters, numbers or words and asked them to repeat the items in order
- He found that the average capacity to repeat the items without any errors was 7
- He also found that some people had capacities of as few as 5 items and as high as 9 items
- Miller referred to this capacity limit as “the magical number 7 plus or minus 2”
Research suggests that when we read things for memorization, the information gets turned into what?
An auditory code
Is auditory STM memory capacity fixed in someone’s brain or does it change throughout someone’s life?
- Research has shown that a person’s memory capacity remains stable across multiple tests & even over many years and thus appears to be a fixed feature of a person’s brain
- Miller’s idea of “the magical number 7 plus or minus 2”
- The plus or minus 2 represents the fact that some people have stable capacities that are higher than others and “magical number 7” is the average
What’s the difference between “low capacity” and “high capacity” individuals for auditory STM memories?
- “Low capacity” individuals are only able to reliably repeat 5 items
- “High capacity” individuals are able to reliably report up to 9 items
What did Miller note measuring capacity in terms of digits or letters (what he called “bits”) doesn’t account for?
He stated that it doesn’t capture the real capacity limits of STM
What’s a chunk?
- Any combination of letters, numbers, or sounds that constitute a meaningful whole
- Grouping items together in a meaningful way so more information to be represented at one time
- It appears that what we can remember in STM are meaningful groups of information, what Miller referred to as a chunk
- The capacity to chunk information into larger units depends on engaging long-term memory, where patterns of previously encountered items (i.e. chunks) are stored
- Once a combination of items is matched to memory, it no longer needs to be represented as separate items but as a single chunk
- Chunking increases with knowledge
Describe how measures of visual short-term memory (VSTM) use a change-detection task
- First, participants are shown a screen with several objects, such as colored squares
- This screen disappears and a new one is shown, which may be identical to the original or have some property changed
- When the number of items in the set is less than 4, performance is nearly perfect but after 4 items, it begins to decline rapidly with number of items in the set
- Using this kind of task, researchers have concluded that VSTM has a capacity of 3-5 items
Describe visual stimuli chunking
- Some research has provided support for a kind of visual chunking, similar to that which takes place for auditory stimuli
- In the case of visual stimuli, people can remember 4 visual objects, even if each object has multiple features, such as a shape, a color and an orientation
Describe how Luck and Vogel (1997) tested visual memory
- They used a paradigm reminiscent of change blindness, where Ps had to identify which square changed colors, in order to test visual memory
- They found that once the number of objects increased to greater than 5, memory performance rapidly dropped off
Describe Inoue and Matsuzawa’s (2007) study on VSTM
- They provided chimpanzees and human participants with 5 digits that appeared on a screen from 210 ms up to 650 ms
- After that wait period, the digits disappeared and were replaced by blank squares
- The subject was tasked with touching the squares to reveal the digits in numerical order
- They found that on average, most humans’ memory performance for the 5 digits decreased when the presentation time dipped below 400 ms
- However, some chimpanzees were able to still store the 5 digits with only 200 ms exposure to the numbers
The ability to chunk information can sometimes give the appearance of what
A super-memory
What are mnemonists?
- People who are capable of memorizing long strings of letters or numbers
- Mnemonists aren’t thought to be skilled due to an enhanced memory capacity but rather due to a skilled ability to form large chunks
- Many mnemonists report that they don’t have an extraordinary memory, some even report having a bad memory in their everyday lives, but they employ memorization tricks that allow them to create consistently larger chunks
Describe Chase and Simon’s (1973) chess memorization study
- They aimed to examine the effect of expertise in the visual domain, comparing chess experts and novices’ ability to memorize configurations of chess pieces on a chess board
- They found that chess experts were able to remember the positions of ~16 pieces while the novices could only remember ~4 (consistent with standard VSTM)
- This doesn’t mean that chess players have larger VSTM capacity
- The effect of expertise was only found to apply when the chess pieces consisted of real configurations that could occur in a chess game
- When the pieces were placed randomly in a configuration that could never occur, experts were no better than novices at recalling the configuration
- This suggests that experts’ greater capabilities in the first case likely depended on combining certain configurations of pieces together into meaningful chunks
- Experts use knowledge of moves to ‘chunk’ pieces together
What did Chase and Simon (1973) demonstrate through their chess memorization study?
They demonstrated that attaching meaning to the information facilitates the storage of information
Describe the Brown-Peterson task
- Consists of an STM task in which participants were told, on each experimental trial, to memorize 3 letters (i.e. a trigram)
- However, immediately after the presentation of the 3 letters, the participants were presented with a 2-digit number and had to begin counting backwards by 3’s out loud, a task that was intended to prevent rehearsal of the trigram
- After a certain duration of counting backward, participants had to report their memory of the trigram
- They found that, under these conditions, the memory of the trigram began to fade after a few seconds and by 15-18 seconds, participants showed little to no memory of the original trigram
What did the Brown-Peterson task results lead Peterson and Peterson to conclude?
- That the duration of STM, in the absence of rehearsal, is ~15 secs
- They attributed their findings to decay and not interference, based on the fact that participants weren’t asked to remember anything else besides the letters on each trial
What are the 2 hypothetical explanations for why memory fades with time?
- Decay models
- Interference
- Some researchers have argued that there’s no decay at all in the absence of interference
Describe the decay models hypothetical explanation for why memory fades with time
- According to decay models, forgetting occurs simply because of the passage of time -> info will just fade away
- According to this view, memory may be thought of like a leaky bucket in which information trickles out
Describe the interference hypothetical explanation for why memory fades with time
- Where new information that comes into memory serves to displace older information
- Forgetting from interfering information
What are the 2 types of interference?
- Proactive interference
- Retroactive interference
What’s proactive interference?
- ‘Forward in time’
- A phenomenon in which information encoded at an earlier point in time interferes with the ability to recall information encoded at a later time
- Learned information causes you to forget something that you learn in the future
- Previously learned material interferes with new information
- Ex: trouble learning a new phone number
What’s retroactive interference?
- ‘Backward in time’
- A phenomenon in which information encoded at a later point in time interferes with the ability to recall information encoded at an earlier time
- Newer information causes you to forget something from the past
- Newly learned info interferes with old info
- Ex: trouble remembering your old phone number
How has the reanalysis by Keppel and Underwood (1962) argued that interference may be at play in the Brown-Peterson task?
- First, the counting task requires a participant to remember each previous number, which could generate retroactive interference of the original trigram
- Also, there might be interference from long-term memory, specifically as participants performed multiple trials over the course of the experiment, it’s possible that some of the letters from previous trials could interfere (proactively) with the current trial
- They also found that in the first few trials, participants remembered the trigram for up to 15 secs with relatively high accuracy. Only on later trials, after the participants had seen many previous trigrams, did the decay over time begin to appear. Suggesting that both interference and decay are at work in the loss of memory in STM
Describe Lewandowsky et al. (2004) study on the decay and interference theories of memory
- They had participants report remembered sequences of letters by responding on a computer keyboard
- They had participants type the sequence at varying speeds, so they could measure the effect of time on recall
- As they performed this memory task, they had to repeat an irrelevant word out loud, a technique called articulatory suppression
- They found that there was no effect of speed at which participants had to repeat the letters on their recall of the items
- They concluded that the passage of time alone, without interference, doesn’t cause decay in STM
What’s articulatory suppression?
A technique used in verbal memory experiments, designed to block rehearsal without causing interference in STM, in which the participant repeats a task-irrelevant utterance out loud while trying to maintain other verbal items in memory
Describe Baddeley’s (1974) working memory (WM) model
- In this model, STM is an active workspace where information could be mentally manipulated based on the current task
- This model proposed that STM isn’t a single unitary store but consists of 3 connected but distinct sub-units
- A critical aspect of WM is that it differentiates between visual and auditory stores -> this division is based on experimental research that suggests that these different modalities really are stored in terms of distinct sensory codes (there’s data that show that auditory memory really does seem to be encoded in terms of auditory properties)
- In the WM model, the visual and auditory buffers are separate from each other and therefore don’t interfere with one another
- WM has a limited capacity and duration
- Where incoming information can relate to prior knowledge and be manipulated
- Where information enters consciousness and awareness
- Critical for LTM formation
What’s the visuo-spatial sketchpad?
- The visual component of the working-memory model that both holds visual information and allows for active manipulation and analysis of that information (manipulation -> ex: mentally rotating a remembered subject)
- Comprised of the visual cache and the inner scribe
- Related to visual semantics
What’s the phonological loop?
- The auditory (or verbal) component of the working-memory model in which information can be repeated/rehearsed
- Similar to the original conception of STM
- Comprised of phonological store and articulatory control loop
- Related to language
What’s the central executive?
- AKA gatekeeper
- Manages and maintains
- Component of the working-memory model that determines what information makes it into memory and toggles between the visual and auditory memory stores
- Component that’s completely novel to the WM model
Describe Conrad (1964) study on “acoustic confusions”
- He found that when people had to memorize a list of letters, they had more difficulty doing so when the letters sounded more similar to one another (ex: “c”, “t”, “v”, “b”) than when they didn’t (ex: “c”, “s”, “r”, “m”, “l”)
- He referred to these errors as “acoustic confusion”
Describe Brooks’ (1968) experiment on auditory-memory tasks and visual-memory tasks
- On each trial, participants had to either do an auditory-memory task or a visual-memory task
- The auditory-memory task consisted of hearing a sentence
- On auditory trials, participants had to then report whether each word in the sentence was a noun or not, in the order in which it appeared
- On visual trials, they had to mentally move along the vertices of the block letter and report whether each one was an “extreme” or not, meaning whether it was a point on an outside or inside corner of the letter
- Brooks employed 2 different ways of responding to these tasks
- In the verbal task, participants could verbally respond “yes” or “no” and in the visual task they would point to a “Y” or a “N” on a computer screen
Describe Brooks’ (1968) findings for his experiment on auditory-memory tasks and visual-memory tasks
- He found that people did better on the task when they had to respond in a different modality than what they had to remember
- Ex: if they were doing the auditory-memory task, they performed better if they used the visual pointing task and vice-versa for the visual memory task
- These results suggest that visual and auditory memory are processed separately from one another and that they don’t interfere with one another, at least not as much as same-modality information does
- His experiment illustrates how subjects completed memory tasks primarily taxing either the phonological loop or visuo-spatial sketchpad and then used either an auditory or visual mode to respond
The central executive in the working memory model plays what 2 roles?
- Coordinating between the phonological loop and the visuo-spatial sketchpad
- Determining what information makes it into STM in the 1st place
Recent research has suggested that individual differences in STM capacity (i.e. the fact that some people can remember more items than others) may be due to what?
To differences in the ability to filter out irrelevant information from memory
Describe Vogel et al. (2005) study on memory capacity
- They gave participants a standard STM task
- Based on the participants’ performance, the researchers divided people into categories of high or low memory capacity
- Both groups then performed a specialized visual STM task
- On all trials, participants were first shown a cue to indicate which items would be relevant to remember
- On standard trials, participants only had to remember the left or right side of the memory display
- Once that slide disappeared, a test display soon appeared and participants were required to indicate whether the relevant items had changed from the memory display
- On some trials, they also presented blue rectangles along with the red ones; participants were told to ignore these and only remember the red rectangles
- After the delay period, when participants would have been trying to hold the information in memory, Vogel et al. used an EEG to measure an electrical event-related potential (ERP) response known to be active during STM activation
- They then compared the ERP responses of the high and low-capacity memory participants
Describe the findings for Vogel et al. (2005) study on memory capacity
- They found that the ERP response was similar for the 2 groups of participants when only the red rectangles were present
- However, when the blue distractor rectangles were present, low-capacity individuals showed a stronger ERP response than high-capacity individuals
- They suggested that this means that low-capacity individuals couldn’t successfully filter out the blue rectangles, even though they weren’t relevant to the task
- This suggests that these individuals’ central executive may not be as effective as those with higher capacity
What did Vogel et al. (2005) findings for their study on memory capacity suggest?
Their findings suggested that a critical role of the central executive is to make sure irrelevant and unwanted information doesn’t enter into memory because it could interfere with the information the system actually wants to retain
Some researchers believe that deficits in memory associated with aging may be due to what?
- To an inability to filter out unwanted information from the working memory
- Hence, age-related memory decline may be due to a decline in the central executive rather than the memory stores themselves
Describe examples of phenomena that challenged the idea that the visual and auditory memory components are completely separate from each other and other system such as LTM
- People can remember many more items when they form a coherent story, a phenomenon related to Miller’s “chunking”
- This indicates that the phonological loop must somehow interact with LTM
- Also, conscious experience appears to bind together different modalities
- Ex: we remember a person speaking with a given voice
What component to his WM model did Baddley add, following the several lines of research that challenged the idea that the visual and auditory memory components are completely separate from each other and other system such as LTM
Based on these and several other considerations, he updated his WM model with a new component called the episodic buffer
What’s the episodic buffer?
- A component proposed as a revision to the original WM model that can combine information from across different sources including the phonological loop and the visuo-spatial sketchpad as well as long-term memory (integrates info from STM and LTM)
- This component is seen as a separate, time-limited, memory store that can combine information from across different sources
- The buffer is controlled by the central executive and is assumed to lead to conscious awareness of time-based, multi-sensory memories
- Related to episodic LTM
Differences in working-memory capacity may be predictive of (& more controversially causally related to) what?
- Other measures of cognition including general intelligence
- Conway (2003) reviewed research on measures of general intelligence and found that they were strongly correlated
What have people linked the correlation between working-memory capacity and general intelligence to?
Some have suggested that this is likely due to developmental factors and children with greater memory spans may be able to increase other areas of cognition
Describe the study by Li & Geary (2017) on the linking of the correlation between working-memory capacity and general intelligence to developmental factors
- They presented evidence for this for the visuo-spatial working memory capacity
- They found that 1st graders’ skills in completing tasks dependent on visual working memory were predictive of their mathematical abilities in 5th grade
Describe the findings of Kali (2007) on the linking of the correlation between working-memory capacity and general intelligence to developmental factors
They found that assessments of memory span administered to children at an early age can predict other measures of reasoning ability later in life
Give examples of findings on the relationship between working memory capacity and general reasoning and intelligence
- Gui et al. (2018) found that short-term musical training can lead to selective improvements in WM ability in young children
- Training to play an instrument for 6 weeks has been shown to improve working memory capacity for maintaining digits
What’s the “brain training” industry?
- Such training often consists of repeated participation in increasingly difficult WM tasks with the idea that if people can increase their WM capacity, this may enhance other areas of cognition as well
- While it’s clear that performance on specific WM tasks improves with training, the evidence is much more unclear as to whether there’s an effect on any other measure of memory or general intelligence
- To date, the evidence suggests that a “buyer beware” approach is warranted
Describe what the working memory looks like at the neural level and what brain areas are linked to it
- It consists of the initial perceptual encoding of information continuing on for as long as the information is being actively remembered
- Working memory happens all over the brain
- Ex: visual working memory engages visual regions of the brain, such as the occipital lobe, while auditory working memory engages auditory areas, such as the temporal lobe
- There’s no one place we can point to in the brain that’s responsible for working memory
- While working memory clearly involves modality-specific regions, the frontal lobes seem to play a particularly important part
Describe Funahashi et al. (1989) study on the brain mechanisms involved in working-memory
- They identified a neuron in the pre-frontal cortex that responded when a square was presented somewhere in the visual field
- They presented this square to the monkey and then made it disappear, followed by a delay period
- After the delay period was over, they gave the monkey a cue to move its eyes to where the square had previously been (requiring that the monkey hold the location in memory)
- They found that the neuron that had responded to the original presentation of the square went into overdrive during the delay period when the monkey had to remember its location
- However, if the monkey wasn’t given the memory task, this response disappeared
Describe Schon et al. (2008) study on the brain mechanisms involved in working-memory
- They used a delayed-match-to-sample task in human subjects
- Using fMRI, they found increased activity in the prefrontal cortex during the delay phase, suggesting an important role of the frontal cortex in short-term/working memory
What the delayed-match-to-sample task?
- Task designed to test visual short-term memory
- Participants are shown an image, followed by a delay and then a second image and are tasked with determining whether the 2 images are the same or different
Describe Moore et al. (2012) findings for their study on the brain mechanisms involved in working-memory
- They found that when regions of the dorsolateral prefrontal cortex are damaged, patients show impairment on tasks similar to the delayed matching tasks, adding more direct evidence for the role of this brain region in short-term/working memory
- Because the frontal lobes are typically associated with executive functions, it’s possible that this area is the site of the central executive, which serves as the “master” over the other cortical areas as they retain perceptual information
What did Stephen Hawking define intelligence as?
As the ability to adapt to change
How did Albert Einstein explain intelligence?
“The true sign of intelligence is not knowledge but imagination”
What’s intelligence?
- The ability to generalize memories (Memory); flexibly use knowledge (Concepts) to solve new problems (Problem solving)
- Ex: I know how to fold a burrito, but how do I swaddle a baby? -> apply burrito folding techniques to swaddling
- Applying the knowledge from other lessons to solve new things
- Thinking and reasoning abilities beyond algorithms
What are the main characteristics of intelligence?
Intelligence relates to efficient and appropriate reasoning:
- Learning from experience
- Adapting to the environment
- Acting purposefully
It varies across individuals
- IQ tests have been designed to measure general intelligence differences
- Other factors underlie differences on these tests aside from “intelligence”
What’s ChatGPT?
- Form of AI
- ChaptGPT solves ambiguous decision-making and reasoning tasks similarly to humans
- ChatGPT uses machine learning to produce content and interact with individuals
- ChaGPT is a set of algorithms that you feed a bunch of words to and it’ll determine what comes next
What problem-solving error does chatGPT fall for, just like humans?
- The conjunction fallacy
- Study where they gave ChatGPT a bunch of decision-making problems and asked it to solve them
- Gave it the Linda problem which highlights the conjunction fallacy
Ex: 1.Linda is a bank teller.
2.Linda is a bank teller and is active in the feminist movement. - ChatGPT just like humans engaged in the conjunction fallacy when trying to solve the Linda problem
What can artificial intelligence do?
- Automated tasks
- Routine activities
- Create content (ex: create Bizarre songs and co-create movies with AI)
What can artificial intelligence not do?
- Editing: avoid repetition in content
- Write accurate news articles (provides ‘fake news’)
- Can’t provide opinions or advice
- Can’t create original puzzles
What’s psychometrics?
The study of psychological assessment
How do we measure intelligence?
With a standard test
What are the characteristics of a standard test?
- Standardization: test scores are compared to pre-tested ‘standardization’ or
‘norm’ groups. A person’s score on some test is calculated by comparing it to a group of people who took the test before - Normal distribution or curve: a symmetrical bell-shaped curve that describes test score distribution. The scores on these tests are thought to follow a normal distribution
Describe IQ tests scores
- IQ scores are said to be normally distributed across the population
- IQ tests tend to be overused
- Average score of 100
- SD = 15
- Within 1 SD of the mean, IQ scores between 85 and 115 (68% of people)
- Within 2 SD of the mean, IQ scores between 70 and 130 (95% of people)
- The top 2% = scores above 130
- Bottom 2% = scores below 70
- Top 0.1% = scores above 145
- Below 0.1% = scores below 55
What’s the reliability of IQ test scores?
- IQ scores have high test-retest reliability
- Scores are the same across testing
- Evidence: Score at age 6 correlates with scores at age 18
What’s reliability?
A consistency in scores from one assessment to the other (across instances of testing)
What’s test-retest reliability?
When you can retake an assessment at different times and get the scores
What’s the validity of IQ test scores?
- IQ scores should have predictive validity if they predict performance on something requiring intelligence
- These tests predict something that requires intelligence (academic achievements and job performance)
- Correlations of 0.5 with job performance
- But what is considered intelligence will vary across context and culture (IQ tests might not be valid everywhere in the world)
What’s validity?
The test is measuring what it is intended to measure
Who and what marked the start of intelligence testing?
- Francis Galton (1822 to 1911)
- Developed intelligence tests, where he looked at reaction times speed
- He was really interested in differences in individuals and their ability to complete cognitive tasks
- His purpose was questionable
- Founded the eugenics movement
- Racially-motivated view of how to
“improve” society - A dark start to intelligence testing
Describe Alfred Binet’s contribution to intelligence testing
- Developed a test in response to a request from the French government
- Purpose to identify children that needed special education in school
- Binet viewed intelligence as important for: practical life, adapting to circumstances, judging and reasoning well
- Binet thought his test only measured academic output and not intelligence