Executive Functions and Working Memory
http://www.psychologicalscience.org/journals/ps/19_9_inpress/Persson.pdf
This is a link to a research article arguing that training on WM tasks can build a cognitive skill called "interference resolution."
The researchers trained subjects on a couple different WM tasks and they biased the distribution of cues to make the tasks harder. They presented cues in clusters so that the incorrect answer on a given trial would often be the correct answer from the last trial and/or the correct answer for the next trial.
At the end of the training program, subjects who trained at the modified WM tasks were faster, than those who trained at a normal WM task with randomized cue distribution, at coming up with unique verbs for animals. (e.g. duck - waddle, kangaroo - hop). Since none of the subjects had trained at the naming task (or anything like it) these results are evidence for far transfer. Since naming speed is be constrained by one's ability to pick the best word from those available - subjects are thought to have strengthened their general capacity to resolve interference.
Interference resolution is thought to be an important part of memory function, and it may play a role in attention.
This could be an idea for a new variation to on cognitive fun game. Or it could be something someone else might want to pick up.
This is a link to a research article arguing that training on WM tasks can build a cognitive skill called "interference resolution."
The researchers trained subjects on a couple different WM tasks and they biased the distribution of cues to make the tasks harder. They presented cues in clusters so that the incorrect answer on a given trial would often be the correct answer from the last trial and/or the correct answer for the next trial.
At the end of the training program, subjects who trained at the modified WM tasks were faster, than those who trained at a normal WM task with randomized cue distribution, at coming up with unique verbs for animals. (e.g. duck - waddle, kangaroo - hop). Since none of the subjects had trained at the naming task (or anything like it) these results are evidence for far transfer. Since naming speed is be constrained by one's ability to pick the best word from those available - subjects are thought to have strengthened their general capacity to resolve interference.
Interference resolution is thought to be an important part of memory function, and it may play a role in attention.
This could be an idea for a new variation to on cognitive fun game. Or it could be something someone else might want to pick up.
godel2 | 3 years ago
Reply
Psychol Sci. 2008 Sep;19(9):881-8.Click here to read Links
Gaining control: training executive function and far transfer of the ability to resolve interference.
Persson J, Reuter-Lorenz PA.
University of Michigan, Ann Arbor, MI, USA.
Functional brain-imaging data document overlapping sites of activation in prefrontal cortex across memory tasks, suggesting that these tasks may share common executive components. We leveraged this evidence to develop a training regimen and a set of transfer tasks to examine the trainability of a putative executive-control process: interference resolution. Eight days of training on high-interference versions of three different working memory tasks increased the efficiency with which proactive interference was resolved on those particular tasks. Moreover, an improved ability to resolve interference was also transferred to different working memory, semantic memory, and episodic memory tasks, a demonstration of far-transfer effects from process-specific training. Participants trained with noninterference versions of the tasks did not exhibit transfer. We infer that the transfer we demonstrated resulted from increased efficiency of the interference-resolution process. Therefore, this aspect of executive control is plastic and adaptive, and can be improved by training.
gold.
pure, unadulterated... gold.
far transfer cognition enhancing "games" will constitute a revolution for education and, probably someday, humanity.
brain like a muscle.
Gaining control: training executive function and far transfer of the ability to resolve interference.
Persson J, Reuter-Lorenz PA.
University of Michigan, Ann Arbor, MI, USA.
Functional brain-imaging data document overlapping sites of activation in prefrontal cortex across memory tasks, suggesting that these tasks may share common executive components. We leveraged this evidence to develop a training regimen and a set of transfer tasks to examine the trainability of a putative executive-control process: interference resolution. Eight days of training on high-interference versions of three different working memory tasks increased the efficiency with which proactive interference was resolved on those particular tasks. Moreover, an improved ability to resolve interference was also transferred to different working memory, semantic memory, and episodic memory tasks, a demonstration of far-transfer effects from process-specific training. Participants trained with noninterference versions of the tasks did not exhibit transfer. We infer that the transfer we demonstrated resulted from increased efficiency of the interference-resolution process. Therefore, this aspect of executive control is plastic and adaptive, and can be improved by training.
gold.
pure, unadulterated... gold.
far transfer cognition enhancing "games" will constitute a revolution for education and, probably someday, humanity.
brain like a muscle.
medicalstudent | 3 years ago
Reply
"Although the present evidence for positive transfer effects may seem to conflict with our prior report of negative transfer (fatigue) following extensive within-session practice (Persson et al., 2007), we view these results as complementary. Strength training provides an appropriate analogy (Muraven, Baumeister, & Tice, 1999). Physical training leads to short-term (i.e., muscle) fatigue, but long-term improvement in strength, provided that the duration of training is sufficient and recovery between training sessions is permitted. Likewise, a specific cognitive process (i.e., resource) might be temporarily fatigued (i.e., depleted) following extensive short-term work, but given sufficient recovery and repeated engagement, this process may operate more effectively in the long term. Indeed, further work is needed to distinguish the parameters that are optimal for negative transfer of the type we reported previously (Persson et al., 2007) from those that are optimal for the positive transfer reported here. We think that distributed and repeated practice, associated with depletion and ample recovery, will prove essential for positive transfer."
michigan seems to like the brain:muscle analogy.
michigan seems to like the brain:muscle analogy.
medicalstudent | 3 years ago
Reply
what does m-i-l-k spell?
what does m-i-l-k spell?
what does m-i-l-k spell?
what does m-i-l-k spell?
what do cows drink?
what does m-i-l-k spell?
what does m-i-l-k spell?
what does m-i-l-k spell?
what do cows drink?
medicalstudent | 3 years ago
Reply
cevapcici | 3 years ago
Reply
I hadn't paid enough attention to that last line you quote. It's interesting to think about the significance of the 4 different aspects of the muscle analogy that they identify:
repeated practice - muscle growth requires repeatedly straining the same muscle groups
depletion - muscle growth requires that one repeat until fatigued
ample recovery - one must rest and consume the proper nutrition in the interim between each practice
distributed - the same exercise cannot be repeated over and over indefinitely - new variations in the exercise have to be introduced or else growth hits an asymptote.
If the distributed part of the analogy holds, then there should see a point of diminishing transfer whenever someone trains intensively on the same game.... So, there would be a limit to how much gain one could hope to get out of, say, training dual N-back at the J-B task.
repeated practice - muscle growth requires repeatedly straining the same muscle groups
depletion - muscle growth requires that one repeat until fatigued
ample recovery - one must rest and consume the proper nutrition in the interim between each practice
distributed - the same exercise cannot be repeated over and over indefinitely - new variations in the exercise have to be introduced or else growth hits an asymptote.
If the distributed part of the analogy holds, then there should see a point of diminishing transfer whenever someone trains intensively on the same game.... So, there would be a limit to how much gain one could hope to get out of, say, training dual N-back at the J-B task.
godel2 | 3 years ago
Reply
there is probably a limit that can be approached [y=x^(1/2)] if one does not fundamentally alter that which governs this limit
but unlike physical training (bodybuilding), cognitive enhancement leads to an improved capacity to understand (and further improve) cognitive enhancement
you may have to get to the limit to learn/understand how to break through it.
but unlike physical training (bodybuilding), cognitive enhancement leads to an improved capacity to understand (and further improve) cognitive enhancement
you may have to get to the limit to learn/understand how to break through it.
medicalstudent | 3 years ago
Reply
y=x^(1/2) doesn't approach a limit. It goes up forever, just slowly.
I still understand what you mean though. (Considering this, I hope my comment was not completely uncalled-for.) A better function would probably be something like y=-1/x + [final gF]
I still understand what you mean though. (Considering this, I hope my comment was not completely uncalled-for.) A better function would probably be something like y=-1/x + [final gF]
LaGrange123 | 3 years ago
Reply
the derivative approaches zero... i was talking about a different limit (intelligence gains per session versus intelligence)
warranted post, good point.
warranted post, good point.
medicalstudent | 3 years ago
Reply
Over-training the muscles can have permanent negative consequences. I wonder if there is a similar risk with mental training...
cevapcici | 3 years ago
Reply
training muscles causes physical (ultrastructural) damage and concomitant elevation of cytosolic enzymes in the blood
i would tend to think it is unlikely mental activity does the same
executive functions are limited by supplies of metabolic resources, training likely depletes theses resources (ATP, creatine phosphate) and the body *may* overstock energetic metabolites in response to training
i am unaware if a)this happens and b)what effect it might have on the gains made in response to training
i would tend to think it is unlikely mental activity does the same
executive functions are limited by supplies of metabolic resources, training likely depletes theses resources (ATP, creatine phosphate) and the body *may* overstock energetic metabolites in response to training
i am unaware if a)this happens and b)what effect it might have on the gains made in response to training
medicalstudent | 3 years ago
Reply
I read a study not long ago which purported to show that people of erudition lose brain volume at a swifter rate than those who are less well-educated. The normal rate of loss for an adult is 2% or so a year and the rate of brain volume loss for intellectuals is around 3-5% (? really not sure about those numbers). (I'm sorry. I can't cite my sources)
I wonder whether or not this is due to glutamatergic excitotoxicity... It stands to reason that the minds of at least some highly intelligent people or those who make greater use of their minds might feature higher baseline levels of glutamatergia.
Damn glutamate... and dopamine.
Is the damage caused by Ca+2 ions influx into the neuron unalienable from its effect of occasioning actions potentials or might there be some way to increase the activity of the brain through glutamate pathways while diminishing Ca+2 ions' harmful effects upon the brain - say by inhibiting proteolytic enzymes and phospholipases?
I wonder whether or not this is due to glutamatergic excitotoxicity... It stands to reason that the minds of at least some highly intelligent people or those who make greater use of their minds might feature higher baseline levels of glutamatergia.
Damn glutamate... and dopamine.
Is the damage caused by Ca+2 ions influx into the neuron unalienable from its effect of occasioning actions potentials or might there be some way to increase the activity of the brain through glutamate pathways while diminishing Ca+2 ions' harmful effects upon the brain - say by inhibiting proteolytic enzymes and phospholipases?
? | 3 years ago
Reply
I get annoyed by the near-universal belief out there that bigger brains, more neurons, and more synapses is always better. In many cases, the opposite is true. We had the largest number of neurons in our brains when we were about six; shortly after that, about half of them committed suicide, and their numbers have continued to fall since then. Have we been suffering from cognitive decline since we were six? I think not. In the mouse barrel cortex (somatosensory cortex for whiskers), sensory deprivation (by shaving the whiskers) *increases* synapse density. Use of lithium causes an increase in the thickness of gray matter in the cortex within a few months, but it also impairs memory, slows cognition, and reduces verbal fluency. (It also makes people less likely to kill themselves or rob banks.) I'm not saying that less is always better in the brain--we all know the story of the London cabbies--I'm saying that it's sometimes better.
If you want to make a wooden sculpture of an elephant, you start with a block of wood and then remove everything that doesn't look like an elephant. If you want to make a neural network that processes information efficiently and accurately, one way is to form a bunch of connections semi-randomly and then prune off the ones that transmit irrelevant, redundant, or inaccurate information. The fewer synapses it uses, the less space and energy it will consume; the less space it uses, the shorter the distances between modules, and the faster the processing and information flow.
I saw a poster at a recent conference where the author claimed that part of glutamatergic neurotoxicity was due to glutamate binding to EAAT1 (the glutamate and aspartate transport protein found in neurons) and blocking (IIRC) aspartate uptake; the lack of aspartate reduced synthesis of some protein that reduced reactive oxygen species, or something along those lines. We'll see if it passes peer review. Anyway, don't be too quick to assume that Ca2+ mediates excitotoxicity.
Also, keep in mind that IQ is positively correlated with longevity, not negatively. I think that also holds true for cognitive decline, though I'm not sure.
If you want to make a wooden sculpture of an elephant, you start with a block of wood and then remove everything that doesn't look like an elephant. If you want to make a neural network that processes information efficiently and accurately, one way is to form a bunch of connections semi-randomly and then prune off the ones that transmit irrelevant, redundant, or inaccurate information. The fewer synapses it uses, the less space and energy it will consume; the less space it uses, the shorter the distances between modules, and the faster the processing and information flow.
I saw a poster at a recent conference where the author claimed that part of glutamatergic neurotoxicity was due to glutamate binding to EAAT1 (the glutamate and aspartate transport protein found in neurons) and blocking (IIRC) aspartate uptake; the lack of aspartate reduced synthesis of some protein that reduced reactive oxygen species, or something along those lines. We'll see if it passes peer review. Anyway, don't be too quick to assume that Ca2+ mediates excitotoxicity.
Also, keep in mind that IQ is positively correlated with longevity, not negatively. I think that also holds true for cognitive decline, though I'm not sure.
? | 3 years ago
Reply
Clarification: whisker sensory deprivation increases synaptic density compared to controls, but not overall; what happens is that some synapses are pruned off when the mouse has whiskers, but are left intact without whiskers.
? | 3 years ago
Reply
I'd like to see that study as well. I'm assuming this study has already controlled for exercise, demographics, nature of work, and dietary habits.
cognitivefun | 3 years ago
Reply
id like to see that study
usually i find the opposite to be true... lifelong learning prevents cognitive decline
once the molecular mechanisms underlying the adaptations to training are elucidated; they can be manipulated; we can take the derivative of acceleration.
glutamatergic signaling is at the core of neuronal communication, in particular, the NMDA/D1 axis in the prefrontal cortex has been called "the engine of cognition" by ivy league researchers.
i dont know how this engine ties in with training... yet.
usually i find the opposite to be true... lifelong learning prevents cognitive decline
once the molecular mechanisms underlying the adaptations to training are elucidated; they can be manipulated; we can take the derivative of acceleration.
glutamatergic signaling is at the core of neuronal communication, in particular, the NMDA/D1 axis in the prefrontal cortex has been called "the engine of cognition" by ivy league researchers.
i dont know how this engine ties in with training... yet.
medicalstudent | 3 years ago
Reply
Loss of brain volume does not necessarily correlate with cognitive decline. Those highly educated tend to suffer less from cognitive decline than those who are less educated (the same holds for those who possess a high IQ as well).
? | 3 years ago
Reply
