Slide Studies (Cognitive Neuroscience)

Damasio 1994

AIM - Proposed Somatic Marker Hypothesis - Emotional processes guide behavior, particularly decision-making - "Somatic markers" are feelings in the body that are associated with emotions - According to the hypothesis, somatic markers strongly influence subsequent decision-making - Processed in orbitofrontal cortex, 'convergence zone' from knowledge, emotion and sensory areas IMPORTANT - Can be thought of as 'gut feelings' which guide behaviour on basis of previous similar experiences EXPLANATION DEFICIT - Patients such as EVR (didnt fail any typical EF tests) have deficit in utilisation of 'somatic markers' - Links emotional state with knowledge of the world: EVR unable to link these things, so could not just 'go with his feelings', and faced constant indecision

Otten (2007)

AIM - Does encoding-related activity also differ depending on how memory is probed? Method: Study Phase - Series of words, seen in text form - Incidental encoding task: is object's width greater than it's height? (make a decision) Test Phase - Recognition memory - Two types of retrieval cues (different input modality) > Spoken words > Pictures Results - See qualitative differences at encoding depending on the type of memory probe Conclusion - When you see a word visually, you generally sound them out in your head and imagine what it looks like - These things can be employed at retrieval to get memory back (ie. you look at different facets of the past)

Stevens et al 2005

AIM - Provided evidence for the effects of temporal discounting on value evaluation EXAMPLE - Task: $20 now, or $28 in 30 days? - Objectively, $28 is of more value RESULTS - Human and monkey making choices for water: decreases over time the value of the water - Pigeon making decisions for food pellets: subjective value drops incredibly fast, waiting even 30 seconds for a second food pellet is extremely unattractive CONCLUSION - But the subjective value of money decreases the further away it is temporally: time discounting - Everyone has a different curve for this: how much money is worth to them subjectively over time

Ozonoff et al 2004

AIM - Tested Executive Function of ASD individuals METHOD + RESULTS Planning: Tower of London task - Seems to be a problem in autism at all ages and ability levels - Less efficient strategy use in complex puzzles Flexibility/set shifting: the Wisconsin Card Sorting Task - Problem in ASD at all ages and ability levels - Tend to perseverate, using old rule unless prompted Inhibition: Stroop task - Doesn't seem to be a problem in people with ASD!

Towgood et al 2009

RESULTS - Found that ppts with ASD didn't show impairments on the WCST when they were explicitly informed that the rule had changed CONCLUSION - Suggests that they are not struggling to switch to a new rule (eg. not an EF problem), but that they are struggling to understand that they are supposed to switch to a new rule - Indicates a problem in inferring implicit task demands rather than an EF deficit

Plassman et al 2007

RESULTS - Found that ppts' 'willingness-to-pay' correlated with activity in the ventromedial PFC CONCLUSION - Suggests that value-based decisions have something to do with this area

Bolton et al 1994

RESULTS - Found that the direct family of children diagnosed with ASD were more likely to show higher levels of autistic traits, but not at a clinically diagnosable level CONCLUSION - Suggests that ASD traits are heritable, and was also one of the first studies that suggested that traits should be considered as a spectrum, with ASD sitting at one extreme - Idea that Autism is not qualitatively different from neurotypical people, but that people with Autism show a more extreme form of ASD traits

Adolphs et al 2005

RESULTS - When instructed to pay attention to the eyes, SM becomes normal at recognising fearful faces CONCLUSION - Normally unable to make use of info from the eye region when judging emotions - Maybe she doesn't use the correct processes which aid recognition?

Baron-Cohen et al 1985

AIM - Tested ToM of autistic children METHOD False belief task -Sally and Anne task (based on Wimmer and Perner's puppet paradigm) - Tested 20 ASD children vs Downs syndrome children and controls - ASD children average of 11 years old, Downs 11 years old, controls 4-5 years old - Requires prediction of action on the basis of an attribution of false belief to the other person (not the the basis of own belief or reality) RESULTS - 16 out of 20 (80%) of Autistic children failed the false belief task and pointed to where the marble really was - Compared to 85% and 86% of neurotypical and Downs' syndrome children who got the task right

Freedman et al 2001

EVIDENCE FOR ADAPTING CODING METHOD - Single cell recordings in non-human primates - Monkeys trained to categorise stimuli as 'dogs' or 'cats' - Morphing allowed control over within-class and between-class boundaries ie. some images were more easy to categorise as 'dog' than others, closer to the boundaries - After training, monkeys retrained with new category boundaries - Frontal lobe neuronal activity recorded RESULTS - More than 20% neurons differentiated between dogs and cats - After retraining: SAME neurons now differentiated between new class boundaries CONCLUSION - Frontal lobe cells able to adapt regardless of boundaries between images - Evidence that cells can adapt their functioning to the specific task - In support of Duncan (2001): rather than multiple executive processes, frontal lobe workspace can adapt to perform many different cognitive operations

Bechara et al (1996)

Iowa Gambling Task: Somatic Marker Hypothesis METHOD - Decks A and B contain frequent high rewards, but also high punishments - Decks C and D contain low rewards and low punishments RESULTS - Healthy volunteers > Rapidly learn to use advantageous decks (C and D) > Show large anticipatory GSR (ie. somatic marker) prior to use of disadvantageous decks (A and B) -> emotional arousal telling them it is the wrong decision - Patients with orbitofrontal damage > Often use disadvantageous decks (A and B) at expense of long-term consequences > Showed reduced anticipatory GSR (fail to activate somatic marker?)

Buckner et al 2008

METHOD - Meta-analysis RESULTS - Across several different methodologies, certain areas are consistently identified > Medial prefrontal cortex mPFC > Posterior cingulate cortex/retrosplenial cortex (PCC/Rsp) > Inferior parietal lobule (IPL) - Show nearly complete converegence across 18 data sets - The hippocampal formation (HF) is shown to be involved - Multiple areas functionally correlated with the HF, reinforcing idea that MTL is included in the network - Lateral temporal cortex (LTC) extending into temporal lobe often consistently observed CONCLUSION - DMN comprises set of brain regions that are co-activated during passive tasks and are connected via direct and indirect anatomic projections HOWEVER Other explanations for the anatomy of the DMN: - Methods that have revealed the default network are based on hemodynamic measures of blood flow (eg. fMRI) - It is possible that pattern reflects a vascular pattern > Either by 'draining veins': where active regions achieve blood flow increases at the expense of nearby regions > Or some other poorly understood mechanism of vascular regulation FURTHER SUPPORT FOR DMN - However, DMN has also been identified by measures using resting glucose metabolism - Measure of neuronal activity through glucose metabolism independent of vascular coupling - Identified similar regions as fMRI research

Mazoyer et al 2001

METHOD - Aggregated data (meta-analysis) across 63 adults - Using blocked PET method (comparing active task blocks and passive rest blocks under PET) - Visually and aurally cued tasks vs. passive rest conditions - Asked ppts what they were thinking about during rest states RESULTS - Noted that rest state generally included: > Generation and manipulation of mental images > Reminiscence of past experiences based on episodic memory > Making plans - Further noted that ppts preferentially reported autobiographical memories

Hayner et al 2007

METHOD - Decide which of two tasks to perform (subtract or add two numbers) - Keep intention in mind during a delay RESULTS - Activity in the medial and lateral PFC during delay predicts choice CONCLUSION - Covert goals represented by brain activity

Ingvar 1974

METHOD - First to aggregate imaging of brains during non-directed tasks RESULTS - Found there was consistent activity during rest, particularly in prefrontal areas CONCLUSION - Suggests brain not idle during rest and there may be localised regions more active

Gilbert et al 2008

METHOD - Meta-analysis across large number of tasks involving ASD participants RESULTS - Shows more rostral area of mPFC activated for EF tasks, and more caudal area of mPFC activated by mentalising tasks (in normal ppts) - EF tasks involving switching between internal and external stimuli > ASD and controls show same BOLD activity in EF area > But ASD individuals also show BOLD activity in the mentalising area ANALYSIS - If poor ToM affects EF tasks, we would expect abnormal activity in mentalising areas

Shallice (1978)

METHOD Judgement/Synthesis: Cognitive Estimation Test > Eg. 'how many camels are there in Holland?' RESULTS - People without damage come up with a strategy to make a reasonable guess: constructing a novel plan for how to answer the question and put together pieces of knowledge to make an answer - People with damage tend to come up with wildly wrong estimations

Ekman et al 1969

Universal Emotional States of Faces - 6 Universal emotions 1) Happy 2) Sad 3) Angry 4) Disgusted 5) Surprised 6) Fear - Asked tribesman in Papua New Guinea (little contact to Western world) to make the facial expressions they would to certain situations, and found they were similar to Western expressions

Gilbert et al (2006)

ARGUMENT BY FRITH ET AL 2003 - The mentalising region of the MPFC is engaged when we attend to our own mental states as well as the mental states of others -> contradicts Gilbert work who suggests that area is to do with directing attention to outside external stimuli AIM: - Investigated the functional specialisation within BA10 METHOD - Conducted a meta-analysis of all papers which report some activation in BA10 - Classified them depending on what type of tasks they were: eg. mentalising and multitasking (task switching, including PM) - Mentalising tasks mostly all medial, hardly any lateral - Whereas multi-tasking was mostly shown in lateral areas RESULT - Seem to be able to subdivided the PFC at least to some degree - Prediction accuracy of the type of task performed purely by part of rostral PFC activated was at 74% CONCLUSION - Suggesting there is some form of functional segregation

Gruber et al 2010

AIM - Are these transient brain states something we have control over? METHOD - Used EEG: 24 ppts - Ppts told they would receive 5% of money won on top of standard payment at the end of the task, to make sure they were actually motivated to remember better on higher paying trials - Intentional memorisation task - Cue indicates amount of money earned if word is later remembered (Eg. £2 >>> Table >>> 20p >>> Cat >>> £2 >> Rose) - 5 types of recognition judgements were classified > Remember (confident word was shown before and could make some recollection about it eg. when word was presented, what they thought at the time, what decision they made) > Know sure (confident word was presented before, but couldn't recall details) > Know unsure (thought it had been presented before, but weren't confident) > New unsure (think is new but not confident) > New sure (confident word is new) RESULTS - Greater pre-stimulus activity if the rewards was higher - Activity elicited by the reward cue predicted later memory of a word but only when high CONCLUSION - Prestimulus activity thus not only signals cue-related activity processing but also an ensuing preparatory state - Suggests we can voluntarily increase brain states

Otten et al 2002

AIM - Can brain activity spanning the period between encoding and retrieval predict outcome? METHOD - Block of words - Ppts had to encode in a semantic way (discern meaning) or a phonological way (rhyme - what it sounds like) - Saw a number of words in a few minutes RESULTS - For the semantic task: activity in the left inferior frontal and inferior medial parietal regions predicted how many words were retrieved - consistent signal - In the non-semantic (phonological) task: superior medial parietal regions predicted how many words were retrieved

HIllman et al 2008

AIM - Can one improving Cognitive Functions with Physical Exercise? METHOD > Recruit sedentary older adults > Divide randomly into exercise (aerobic activity) and control (stretching) groups Exercise: - Light to moderate physical exercise - 60-85% of maximal heart rate - At least once a week - Eg: taking the stairs, cycling, dancing, nordic walking, swimming, light rowing > Assess several cognitive functions before and after intervention > No cognitive training RESULTS - Exercise group showed significant benefits on executive function tasks, but also controlled, spatial and speed tasks (in that order)

Hesselmann et al 2008

AIM - Can prestim activity bias perceptual decisions IMPORTANCE - perceptual consequences of endogenous variations in prestim activity remain poorly understood METHOD - fMRI - 12 Ppts shown Rubin's vase-faces picture: can be perceived either as a vase or as two faces - 90 trials of upright Rubin's vase-faces, and 10 inverted - On inverted trials, pictures perceived as vase 2x the amount of times on normal trials -> evidence that ppts were not just making their minds up before the picture was shown, but that they were waiting to see the image before making a perceptual judgment - Focused on the FFA as a region of interest RESULTS - Found that prestimulus activity in the FFA predicted that ppts would be more likely to perceive the picture as two faces - Found no sig. signal deviations in control areas, suggesting the result was not just due to preceding levels of arousal or selective attention CONCLUSION - Suggests that endogenous variations in prestim activity biased subsequent perceptual inferences

Tversky et al 1991

AIM - Demonstrated loss aversion METHOD - Ppts shown coin flip - Asked if they want to choose to flip the coin (Y/N) > Win +$32 vs. Lose -$30 > Win +$32 vs. Lose -$18 > Win $45 vs. Lose -$18 RESULT - Usually have to increase the gain amount by about double the loss amount CONCLUSION - Losses hurt about twice the amount that wins feel good

Tversky et al 1981

AIM - Demonstrated that we dont make rational choices EXPLANATION - Rational choice requires that the preference between options should not reverse with changes of frame - BUT people systematically violate the requirement of consistency, depending on the frame METHOD (part 1) - Asian Disease Problem - Askey to make choices between options which lead to certain outcomes - First choice between A and B (gain frame): > Program A: "200 people will be saved" > Program B: "there is a 1/3 probability that 600 people will be saved, and a 2/3 probability that no people will be saved" -> 72% choose A - Then choose between C and D (loss frame): > Program C: "400 people will die" > Program D: "there is a 1/3 probability that nobody will die, and a 2/3 probability that 600 people will die" -> 78% choose D ANALYSIS - Even though A and C have identical outcomes and B and D have identical outcomes - If people are consistent, they should choose A and C OR B and D CONCLUSION - People are risk averse when the choice involves gains - People are risk taking when the choice involves losses

Tambini et al 2017

AIM - Does an emotional task evoke an emotion-related brain state that lingers and affects later neutral images? METHOD - Used emotional pictures: task blocks of emotional pictures and neutral pictures Condition 1 - Emotional encoding > Rest > Neutral encoding > Rest Condition 2 - Neutral encoding > Rest > Emotional encoding > Rest Condition 3 - Neutral encoding > Rest > Neutral encoding > Rest - Both behavioural and fMRI measures used - Ppts rated the visual complexity of emotional (negative, high arousal) scenes and neutral scenes - Performed memory test 6 hours after scanning session HYPOTHESIS - emotional state would carry forward to neutral state, making it somewhat emotional - An emotional state which affects later processing RESULTS - Memory generally better for images showing emotional info - Skin conductance and arousal level was elevated for neutral stimuli that followed emotional stimuli - Whereas when neutral block came first, there was a significant increase in skin conductance when ppts then saw the emotional block - This suggests that level of arousal carried over - Memory was also elevated for neutral stimuli that followed emotional stimuli - Emotion-related fMRI activity (slow and stimulus-evoked fluctuations) were reinstated for neutral stimuli following emotional stimuli CONCLUSION - Thus, an emotional experience can persist in time and bias how new information is perceived

Otten et al 2006

AIM - Does electrical brain activity in the cue interval predict subsequent recollection of the word? METHOD Study phase - Incidental encoding tasks on a series of words, each preceded by a cue - Cue indicates type of decision (semantic/non-semantic) Eg. o >>> Table >>> x >>> cat >>> o >>> rose - Where o = semantic decision (what does it mean) - And x = non-semantic decision (what does it rhyme with) Test Phase - Recognition memory with confidence judgments - Look at activity (EEG) between presentation of cue and presentation of word: can this activity predict whether you are more likely to remember the word? RESULTS - With semantically encoded memory, you can > Different activity for trials that were later remembered vs. those that were forgotten - Dissociable from activity after an event > Can predict if you will remember a word, even before you are actually shown the word CONCLUSION - Activity before the word is presented can increase encoding efficacy

Gilbert et al 2006

AIM - Feared that mindwandering happened during (easy) stimulus-oriented conditions - Follow up study to check that activity in medial BA10 during stimulus-oriented tasks wasn't just down to mind-wandering during those tasks (because they are easy) METHOD - Similar to Gilbert et al 2005 - BIT also gave them a baseline condition: where screen would flash and had to press button when you see the flash (associated with medial BA10 activity) RESONING > If function of region is mindwandering, then trials with highest activity in these trials should be the ones where RTs were slowest, because you are distracted from the task > But if the region is indeed involved in directing attention to outside stimuli, should see the highest activation for the trials with the fastest RTs RESULTS - Replicated previous findings - Found that signal was greatest on trials with fastest RTs, consistent with the idea that the area is to do with focusing attention on the world around you CONCLUSION - Previous results were not due to mindwandering

Chib et al (2012)

AIM - Follow-up study using Tom et al (2007) task to measure individual correlations in behavioural and neural loss aversion RESULTS - Significant correlation between increasing behavioural loss aversion and striatal deactivation during motor action - The more loss averse the ppts, the greater the ventral striatal deactivation during motor action CONCLUSION - Degree of loss aversion predicts propensity to exhibit performance decrements for large incentives at the hard difficulty level

Galli et al 2011

AIM - How do we encode emotional events METHOD - Incidental encoding task (indoor/outdoor judgements) - Cue indicates valence of upcoming picture (neutral, pleasant, unpleasant) - Emotional pictures matched for arousal Test - Recognition memory test after 20 min delay - Remember/know/new judgements RESULTS - Just before a picture, in response to cue, activity influences how well you will remember it - When items were negative, activity changes to help you remember them better, especially in women

Galli et al 2012

AIM - How do we regulate emotions when taking on different persespectives? METHOD - Incidental encoding task (indoor/outdoor judgements) - Intermixed unpleasant and neutral images - Cue signals valence of upcoming item - Instructions: let feeling arise naturally, or take an outsider position - Women only Test - Recognition memory test after one day - Confidence judgements RESULTS - See prestim activity if emotions arise naturally, but not if you take an outsiders perspective - Seems to have something to do with feeling or regulating feelings to stimuli

Kable et al 2007

AIM - Investigated brain Activity during Temporal Discounting METHOD - fMRI study measuring activity during temporal discounting task - Gave people th option of taking $20 (real money) now, or a greater amount of money at a later date - Plotted the subjective decrease in value of money over time for each ppt RESULTS - Some were more patient whereas others were more impulsive - Found activity in the medial prefrontal cortex (mPFC), the posterior cingulate cortex (PCC) and the ventral striatum - Additionally, found that the amount of activation in these areas corresponded with ppts' subjective valuations -> more willing to wait = higher levels of activity (showing that the amount was of more VALUE to them) CONCLUSION - Shows that the 'patient' ppts are genuinely perceiving the amount as more rewarding/valuable - Shows that the objective amount and delay associated with it alone are not enough to predict behaviour, as it differs across individuals

Chib et al 2012

AIM - Investigated choking under pressure METHOD - Ppts performed a highly-skilled motor task of controlling a virtual spring-mass system - Novel, no previous experience or expertise - Saw visual feedback of targets and hand position > Need to place hand cursor and mass cursor into target within 2 seconds -> difficult task - After 500 trials, ppts get better - On day 2: 6 levels of incentive introduced > $0, $5, $25, $50, $75 and $100 > Two levels of difficulty: easy and hard RESULTS - Most ppts reach peak performance at an incentive level less than $100 - Incentive presentation: BOLD signal in ventral striatum increased with magnitude of incentive - Motor task: BOLD signal in ventral striatum decreased with magnitude of incentive - Ventral striatum activated at time of incentive and deactivated at time of action CONCLUSION - Deactivation of ventral striatum during motor action reflecting evaluation of potential loss (of presumed gain) that would arise from failure to be successful? - Larger incentives framed as larger potential losses?

Li et al 2004

AIM - Investigated cross-sectional age gradients of intellectual abilities IMPORTANCE - Different cognitive functions show different cross-sectional trajectories over the lifespan METHOD - Looked at crystallised intelligence, processing robustness/processing speed and fluid intelligence RESULTS - Peak at around 20 and then decline, but at different speeds and to different degrees - Crystallised intelligence remains more intact - Fluid intelligence and processing speed declines significantly - Children and older adults may show the same performance level - However, this does not reflect comparable brain processes

Caspi et al 2002

AIM - Investigated example of Gene-Environment Interaction of antisocial behaviour METHOD - Children followed from 1972 in longitudinal study - N ~ 1000 - Found that genetic polymorphisms can lead to inter-individual differences in brain function - A single base difference found between chromosomes in the DNA sequence ie. genetic info is referred to as a single nucleotide polymorphism (SNP) - Investigated gene encoding for MAOA enzyme > Codes enzyme that degrades neurotransmitters (dopamine, noradrenaline, serotonin) > Mice with too little MAOA are more aggressive > Knock-out mice deal less well with stress > MAOA polymorphism results in interindividual differences in activity of MAOA enzyme in humans RESULTS - Interaction between low MAOA levels and childhood maltreatment: significant increase in antisocial behaviour when the two interact CONCLUSION - Shows that it is not just your genes but the things which happen in your life, and how you are equipped to deal with those events

Wager et al 2015

AIM - Investigated where emotions may be located in the brain METHOD - Meta-analysis across 148 studies to identify neural patterns that differentiate anger, disgust, fear, happiness and sadness RESULTS - Categories distributed across systems rather than single regions - Consistent with categorical or dimensional theories: responded to both types of emotions as well as intensity of emotions (both seem to affect what systems are activated) - Composites, not individual regions

Senju et al 2009

AIM - Investigated how some ASD individuals can still pass some ToM tests METHOD - Designed for 2 year olds but has also been used as early as 6 months old Non-verbal animation - Bear moving an object from one box to another when the person isn't looking - On true belief trials, the person puts their hand through the window above the box where the ball is RESULTS - Infants look at where they're expecting the hand to come through in anticipation - Adults with ASD look equally between the two windows at chance level, whereas neurotypical adults are significantly above chance for the correct window - NT 2 year olds are also able to pass the task CONCLUSION - ASD adults show a lack of spontaneous mentalising through anticipatory eye movements on the most basic of tasks, despite passing them explicitly - Intact task performance does not equate to intact spontaneous mentalising - Compensations must be more rule-based/explicit

Shing et al 2008

AIM - Investigated if one can maintain a high level of cognitive functioning in ageing? METHOD - Compared children, teenagers, younger adults and older adults - Task designed to improve episodic memory (known to decline in ageing) over several weeks - Learn word pairs of German and Malay language RESULTS - Older adults and children start lower than teenagers and young adults but improve when they are given a strategy and practice - However, all other groups given the strategy and practice improved more - Older adults improved, but not as much as younger people CONCLUSION - In general, transfer to other cognitive functions that have not been trained appears to be limited - Cognitive training also works for older adults (example: episodic memory)

Tom et al 2007

AIM - Investigated loss aversion in the brain METHOD - 16 ppts - fMRI - The bigger the potential loss, the LESS BOLD activity (don't see increase in activity for loss as you might expect): in the mPFC and Striatum - The bigger the potential gain, the more BOLD activity in mPFC and Striatum - Individual differences in behavioural loss aversion are correlated with neural loss aversion (neural activity for losses vs. gains) RESULTS - Ppts that are not loss averse behaviourally -> Similar eural activity for losses and gains - Ppts that are very loss averse: more decrease in neural activity per $ for losses than increase for gains

Bartra et al 2013

AIM - Investigated neural correlates of subjective value METHOD - Meta-analysis: 206 fMRI studies RESULTS - Found that activity consistently reported in: > Ventromedial PFC > Anterior ventromedial striatum - Greater BOLD response for more rewarding (or less aversive outcomes) when a decision is confronted or outcome delivered for monetary and primary rewards CONCLUSION - Maybe that is a 'valuation system'

Castelli et al 2002

AIM - Investigated the Neural Basis of ToM METHOD - 10 ASD and 10 control ppts PET scanned while watching 3 types of triangle animations (mentalising, goal-oriented and random) RESULTS - Found that ASD participants gave fewer and less accurate descriptions of the mentalising animations, but were not sig. diff on the other animations (confirming other research) - ASD individuals showed less activity in certain areas know as the mentalising network when watching mentalising animations: > Medial prefrontal cortex > Superior temporal sulcus > Temporal poles - Did show the same amount of activity in the extrastriate cortex (visual input), but in ASD ppts showed less functional connectivity with other areas CONCLUSION - Authors suggests lack of mentalising in ASD ppts occurred as a result of a 'bottleneck' between lower and higher order perceptual processes (ie. connections between visual cortex and higher processing regions) - Suggests that the brain system that enables mentalising is abnormal in autism > Mentalising system under-activated > Visual system normally activated > Poor connectivity between areas

Hedden et al 2012

AIM - Investigated the age differences in top down control IMPORTANCE: - Older adults have been shown to over-recruit (lateral frontal) areas METHOD Task - 4 conditions increasing in difficulty from 1 to 4 1) Neural nonshift 2) Neural shift 3) Incongruent nonshift 4) Incongruent shift - Looked at: younger adults, older modulating adults, and older non-modulating adults (modulation/non-modulating groups distinguished after testing) RESULTS - Overall activation: the harder the task, the more activation each group shows - Older modulating adults show higher levels of activity even at the easier task, and then show significantly greater levels of activity than the non-modulating group as task gets harder - Non-modulating adults start at similar levels to modulating adults at the easy task, but then activity remains similar - Medial and lateral prefrontal areas heavily activated in the older adults CONCLUSION - Older adults work their PFC as hard as they can rather than recruiting parietal areas - Modulating adults perform better

Rugg et al 2013

AIM - Investigated the brain networks underlying episodic memory retrieval METHOD - Review RESULTS Key retrieval-related brain regions: - Medial Temporal Lobe - Posterior Parietal Cortex - Prefrontal Cortex - Very similar to encoding areas! MTL: - Hippocampus important but also Parahippocampal cortex, perirhinal cortex, rhinal sulcus, entorhinal cortex and hippocampus - Each of these regions thought to play a crucial role in retrieval

White et al 2014

AIM - Investigated the neural basis of the strange stories of triangles (3 types) METHOD - fMRI study using the strange story paradigm 1) Mental stories 2) Animal physical stories 3) Natural physical stories - Paradigm modified so that ppts were asked MCQ's about the stories afterwards, so that it would work in fMRI scanner - 3 groups of adolescents 1) ASD ToM- : poor theory of mind performance 2) ASD ToM+ : good theory of mind performance 3) Neurotypical: matched for age and IQ - Do the ToM+ group have truly intact ToM, or are they somehow compensating? - Used mentalising network as regions of interest (ROI) analysis - mPFC, PCC, bilateral temporal parietal junction/temporal poles RESULTS > Compared to the neurotypical group, both ToM- and ToM+ groups were showing increased activity, but no sig. diffs. between ToM groups > Increased activity = BOLD signal change mental > nature (so they are showing increased activity for the nature stories) - Despite intact ToM performance, ToM+ group show abnormal brain activity consistent with mentalising impairment - ToM impairments universal feature in ASD? CONCLUSION - Lends credence to clinical intuition of unifying specific impairments in autism - Suggests ToM+ are compensating for their neural abnormalities: - Relatively early, spontaneous?

Chang et al 2015

AIM - Investigated where emotions may be located in the brain METHOD - 182 ppts rated their emotional state after viewing negative and neutral scenes, followed by pattern classification analyses RESULTS - Multiple, rather than individual regions predicted negative affect - Network included anterior cingulate, insula and amygdala - Consistent with dimensional theories of emotions

Kounios et al 2006

AIM - Investigated whether being in a certain brain state helps solve a problem HYPOTHESIS: - A distinct type of mental preparation (a certain brain state) would facilitate problem-solving independently of problem-specific knowledge METHOD - Ppts mentally prepare to solve a verbal problem (eg. what connects these three things: bump, egg, and step? answer = goose) RESULTS - Frontal alpha and BOLD activity in the anterior cingulate predicts later problem solving with self-reported insight (the 'aha' moment) - Ie. you could predict whether someone would come up with the right answer, and whether they would report having an 'aha' moment (when they suddenly realised the answer) based on their brain activity before and during CONCLUSION - Something about setting up your brain in the right state to be able to tackle a problem and realising you have come up with the right info

Duzel et al 1999

AIM - Investigated whether being in a certain brain state helps solve a problem Method - EEG for around 20 seconds - With semantic and phonological tasks Results - Shows differentiation of activity for episodic and semantic retrieval - Retrieval is only successful for episodic memory if you tap into/engage these states

Zheng et al 2015

AIM - Investigated whether emotion influences cognition METHOD - Attentional blink task - People see very quick series of images: some emotional, some neutral, some scrambled RESULTS - After emotional stimuli, have a longer attentional blink, where no other stimuli are detected CONCLUSION - So emotional reaction to stimuli can affect attention

White et al 2009

AIM - Investigated whether executive function is affected due to ToM as the primary deficit IMPORTANCE > Analysis of types of mistakes made of EF tests show an unusual pattern in ASD ppts METHODS 6 Parts test - More ecologically valid: 3 types of task (two of each task), have to do something from all tasks in 5 minutes, but can't do the same task twice in a row - Requires multi-tasking ability Zoo Map 2 - Ppt has to travel around zoo map, following certain rules, visiting certain animals at specified times etc - Meant to follow order given by experimenter RESULTS - ASD ppts spent longer than control on any one subtest: controls were more likely to divide time evenly between tasks - ASD ppts said things like: 'I don't want to do much of those ones cos they're boring' - ASD individuals tend to score lower on the Zoo Map test - One child: only visited 5 out of 8 specified animals in order to 'save the others for another day' - Another child visiting the lions out of order because 'I'm a cat lover so I always go there first' - Incorrectly visiting the cafe last because 'I'd be hungry then after walking around the zoo' CONCLUSION - The common denominator is actually poor understanding of the task demands - Poor understanding of the experimenter's expectations - Responds to own desires instead, impose own demands on the task

Lakshminarayan et al 2011

AIM - Is the framing effect something learned with experience of more basic/fundamental? METHOD - Monkeys taught to swap tokens for food: trained that they can give token to human and will get some food in return - Created a 'money market-place' with 'human salesmen' -> could trade tokens with two possible human salesmen - Monkeys learned who sold what, at what price, and how reliable they were - Some salesmen gave more out than others for the same price - monkeys learn which salesmen are better - Also introduced salesmen that were 'safe': gave the same outcome each time - Or ones that were 'risky': gave different amounts - And scenarios where the monkeys got bonuses and losses Bonuses (gain-framed) - Two salesmen: both give bonuses - One safe: gives the same bonus each time (one extra) - One risky: sometimes gives no bonus, other times gives a larger bonus (eg. two extra) Result - Found that monkeys play it safe, like people Losses (loss-framed) - Two salesmen: both give less than expected - Safe loss: takes one away very time - Risky loss: sometimes takes none away, other times takes two away Result - Monkeys become more risky, again the same as people do CONCLUSION - Humans and capuchin monkeys show frame dependent risk preferences - Switch from risk averse (gain-framed scenarios) to risk-seeking (loss-framed scenarios), depending on how the same outcomes are presented - This switch in preferences is irrational: inconsistent choice behaviour - Strong preference to avoid loss: loss aversion

Norman et al 1986

AIM - Propose Attentional Control Model -> basic unit of action control: 'schema' - Schemas activated in 2 different ways: 1) Contention Scheduling: well-known response to stimulus -> Primarily bottom-up automatic triggering in routine situations 2) Supervisory Attentional System (SAS): top-down control in non-routine situations -> Help to overcome automatic situations EXAMPLE Stroop Task - Word-reading schema automatically triggered by presentation of stimuli - But if goal is to read the colour, SAS has to shut down word-reading schema and activate colour reading schema DEFICIT EXPLANATION Model provides account for performance of patients with frontal damage on tests of EF: - Stroop task: once activated, dominant schema 'read word' persists in absence of inhibition by the SAS - Multiple errands task: disruption to selection and appropriate sequencing of schema by SAS leads to disorganised behaviour > Perseveration: schemas fail to be inhibited after operation (eg. Wisconsin Card Sorting Task) > Distractibility: non-task related schemas capture attention > Utilisation behaviour: in absence of SAS inhibition, attention captured by information entering perceptual system

Duncan 2001

AIM - Proposed Adaptive Coding Models - Executive control subserved by common 'global workspace' represented in the frontal lobes (rather than finely sub-divided area) > Frontal lobe neurons flexibly code task-specific information - Can adapt to requirements of a task and perform it: frontal lobe neurons know how to 'load' each task - Exerts top-down control over other processes in the brain to complete task > Similar to the SAS proposed by Norman and Shallice: except there are not different areas/systems which perform distinct functions EVIDENCE - Meta-analysis: found PET and fMRI studies which investigate different aspects of EF QUESTION - Do they activate different parts of the frontal lobe? RESULTS - Results and identified areas all jumbled up: no obvious partitioning where different areas seem associated with particular tasks

LeDoux et al 2000

AIM - Proposed two routes for Sensory Input to the amygdala EXPLANATION - Thalamus and cortex provide inputs to the same neurons in the lateral amygdala - Basic 'dual-route' hypothesis is that the cortex provides a slow, fine grained route for fear, while the thalamus provides a 'quick and dirty' route - This helps: fast reactions; to prime the amygdala for cortical inputs; to allow projections from amygdala to cortex to prime cortical activity - If you see something like a snake, want to act rapidly, not stop and process information via the cortex EVIDENCE - Lesion studies show that the cortex is not required for fear conditioning to basic tones, but may be required for more complex tasks (eg. requiring discrimination)

Gilbert et al 2005

AIM - Testing the Gateway Hypothesis METHOD - Put ppts in MRI scanner and asked them to either pay attention to screen in front of them, or to info internally in their heads - Looked at difference in the brain using fMRI - 2 types of tasks which ppts could accomplish either by using visually-presented info, or by doing the same task 'in their head' - Ppts alternate between the tasks Stimulus-Oriented (SO) Phase: - 1) See a shape and have to navigate around the edge of the shape and press button for the turns you would need to make Stimulus Independent (SI) Phase - 1) Shape turned into thought bubble picture: asked to imagine shape in their head and do the same task but imagined in their head - Looked at what is different in the brain between two types of tasks RESULTS - Regions active in SI task compared to SO task - No significant voxels found - BUT Regions active in SO task compared to SI task = activity in medial BA10 (medial rostral PFC) - Regions active when ppts switched between task compared to when they stayed on same type of task -> Lateral rostral PFC activated - Activity in medial and lateral rostral PFC when co-ordinating the balance between paying attention to internal and external info CONCLUSION - Medial BA10: focusing attention on the external environment - Lateral BA10: switching attention between stimulus oriented and stimulus independent thought - So Rostral PFC supports selection between stimulus-oriented and stimulus-independent thought

Fell et al (2001)

AIM When does the brain start to encode new LTM? METHOD - EEG - Recording directly from MTL in epilepsy patients RESULTS/CONCLUSION - ~400 ms for words and ~200ms for pictures - Overall, about 200ms after you encounter something - And activity can last for several seconds (long-lasting) - typically for around a second and a half

Gilbert et al 2007

AIM - Investigated Mentalising vs. attention functions in medial rostral PFC METHOD - Used original stimulus oriented/independent tasks as in Gilbert et al 2005 - But added additional aspect of mentalising > In mentalising condition, told that switching between conditions is caused by someone else in a different scanner > Some of the time make things difficult for the ppts, and sometimes make it easy > Ask ppt whether the other person was trying to make it easy or difficult for you (mentalising -> because you have to think about the intention of others during the task) > Non-mentalising condition: just told that the task is switching at random by a computer (no need to think about other people's intentions) OVERALL DESIGN - 2x2 factorial design > Stimulus-oriented mentalising > Stimulus independent mentalising > Stimulus-oriented non-mentalising > Stimulus-independent non-mentalising - In interview afterwards, checked whether ppts believed it was true: people seemed to believe there was a difference between the two versions of the task RESULTS - Stimulus-Oriented > Stimulus-independent and Mentalising > Non-mentalising = activated very similar areas of the medial PFC - BUT there was virtually no overlap between the two - Mentalising slightly posterior to the attentional areas (anterior regions activated) CONCLUSION - Provides evidence for meta-analysis: some fine grained functional segregation in this area in the brain

Duverne et al 2010

AIM - Investigated the effects of age on the neural correlates of retrieval cue processing and whether it is modulated by task demands METHOD - EEG - Ppts: young (18-20 years old) and older (63-77 years old) adults - Subjects studied either words or pictures presented in one of two visually distinct contexts, and then performed a yes/no recognition task with words as test items. OR - Subjects made recognition judgments, but were also required, in addition, to signal the study context for each item judged "old." RESULTS - Young subjects' ERPs for new (unstudied) test items were more negative going when the study material was pictures rather than words, and this effect varied little between the two retrieval tasks. - The effects of study material on the ERPs of the older subjects were attenuated and nonsignificant in the recognition task - In the source retrieval task, however, material effects in the older group were comparable in both onset latency and magnitude with those of the young subjects CONCLUSION - Failure of older adults to demonstrate differential cue processing in tests of recognition memory likely reflects the adoption of a specific retrieval strategy rather than the incapacity to process retrieval cues in a goal-directed manner.

Uncapher et al 2009

AIM: - Investigated how the formation of episodic memories is affected by attention during the event processing METHOD: - Review Results: - Activity in dorsal parietal cortex associated with memory success -> Goal-directed attention to-be-remembered/encoded info - Ventral parietal cortex associated with memory failure -> Attention captured by irrelevant info > Reverse: the more activity the less likely you are to remember(because your attention is on something else

Shallice (1982)

AIM: - Investigated planning aspect of executive functions METHOD - Tower of London task -> involves moving away from goal to reach the target RESULT > People with frontal lobe dysfunction struggle to make these counterintuitive moves to reach the goal

Kim (2011)

AIM: - Reviewed whether any single memory encoding system exists METHOD - Meta-analysis of studies considering subsequent memory and subsequent forgetting Conclusion - Almost the whole brain can show these subsequent memory effects - Several key players in memory encoding: > Ventrolateral PFC > Dorsolateral PFC > Posterior parietal cortex > MTL - There is not a single encoding system

Hammerer et al 2010

AIMS - Separate subprocesses in EEG measurements related to performance monitoring: and specifically related to the monitoring (attention) part of the application (motor control) part - Add neurophysiologial measures to understand particular difficulties observed in children and adults METHOD Task - Continuous Performance Task - Instruction: respond as fast as possible when the BLUE square is followed by the YELLOW square - NoGo trials: Every once in a while the blue square is followed by an orange square instead, and ppts have to inhibit their automatic response - 4 age groups: 9-10 years, 13-14 years, 20-30 years and 65-75 years - Measured 4 task-relevant processes using EEG: 1) Attention to cue (blue square): how much attention they pay to the cue 2) Between cue and target stimulus: expressing how certain/focused you are on what you have to do 3) Conflict monitoring: identifying when there is a conflict (eg. when orange instead of yellow square appears) 4) Response inhibition: inhibiting automatic response in favour of correct response HYPOTHESIS - Childhood development: expected deficits in processes related to applying motor control (eg. Geier and Luna, 2009; Jonkman et al, 2006) - Ageing: expected deficits in attention to cue stimuli (eg. Braver et al, 2001; Paxton et al, 2008) RESULTS - Reaction time on Go trials is larger in children and especially older adults (attentional difficulties?) - On 'No go' trials, children and adolescents (motor control problems) make more errors than younger and older adults -> Study controlled for speed-accuracy payoff, so it wasn't just that children were less careful so sacrificed accuracy for being faster CONCLUSION - Older adults show weaker attention to the cue stimulus > They were the only group which didn't show differentiation in activity between cue and non-cue (green square) - Conflict monitoring signals ('no go') are reduced in older adults > Even though children and adolescents made more errors, the EEG signal suggests they were perceiving errors correctly - Response inhibition: children show weaker signals of response inhibition (inhibiting on nogo trials) > Don't show strong differentiation in activity between trials you do and don't inhibit on even though they correctly perceive the conflict - Children show weaker signals of response readiness SUMMARY - Lifespan differences in EEG correlates underlying performance monitoring - During child development, weaker indices of internal motor control (ability to inhibit motor response) are observed - During ageing, internal motor control mechanisms appear to be intact, but a weaker attentional focus on monitored events (attention to cue stimulus) is observed

Wise et al 2004

Critique: Vascular explanations for DMN RESULTS - Demonstrated correlated, spatially specific fMRI responses suggesting that fMRI patterns can reflect vascular responses to breathing CONCLUSION - The spatial patterns associated with respiration do not closely resemble DMN, but are a reminder that a vascular account should be explored further

Yoo et al 2012

IMPORTANCE - Intentional encoding of indoor/outdoor scenes, using the parahippocampal place area (PPA) (known to be involved in spatial processing) AIM - Look at PPA activity while performing task METHOD Exp 1 - Goal of Exp1: identify 'good' and 'bad' PPA learning states - You will never be completely focused on the task all of the time, better at some times than others - So identify what scenes were remembered well at test, and then look back at the activity present at that time to see what 'good' PPA activity looks like Exp 2 - Stimuli presented during 'good' PPA states enhances recognition memory RESULT - When a 'bad' state was detected in parahippocampal place area, ppts did not learn scenes as well

Levy et al 2011

IMPORTANCE - Medial PFC and ventral striatum seem to be the two critical areas where our preferences are constructed AIM - Can you infer choice from brain activity? - Tried to determine whether ppts' choice prefrences could be predicted from brain activity in the mPFC, PCC and striatum METHOD Phase 1 - 12 ppts shown images of objects while fMRI scans brain activity - First shown lotteries where there was an obviously more rewarding choice (ie. $40 vs. $0) to establish where there was activity in the brain for rewarding items - Confirmed that the mPFC, PCC and Striatum were more active Phase 2 - Then ppts viewed goods in the scanner - Just shown the goods ('just-looking' trials) - 226 trials in total - Measured the amount of neural activity in the regions of interest in response to seeing each item - Could then rank the goods in order of most to least activity for each ppt - Predicted: a good with higher levels of activity would be chosen more often than goods with lower activity Phase 3 - Ppts then made choices between goods outside the scanner: a total of 380 choices - Could keep 1 item afterwards -> real incentive RESULTS - - Items that people preferred showed higher levels of brain activity when thy were just being looked at - Neural activity in th mPFC and Striatum was correlated with choice preferences - Correlation between BOLD signal in mPFC and rank preference for good: r = .49 (p<.05) - Correlation between BOLD signal in Striatum and rank preference for good: r = .50 (p<.05) - 80% prediction accuracy when items = 'neurally distant' (ie. when they were further away in rank activation) CONCLUSION - Activity in the mPFC and Striatum is correlated with actual choice preferences - Choice is not necessary to infer values: values are represented in the brain in the absence of choice, and these values can be used to predict later choices - Suggests we are making value judgements even when we are just looking at items (and not yet making a decision)

Gilbert et al 2008

IMPORTANCE - Studies suggest abnormal activation of medial rostral PFC in ppts with ASD during mentalising tasks (Castelli et al, 2002) AIM - What about a nonsocial task that also activates medial rostral PFC? METHOD - fMRI - ASD and control group matched for age and IQ - Both did task where letters in red have to identify whether they have a curve in them (stimulus-oriented phase) - Or the letters are in blue and you have to continue the sequence mentally and ignore what you see on the screen (stimulus-independent phase) RESULTS - Stimulus oriented compared to Stimulus independent: > For both groups see activity in mPFC as expected but not same region - Slightly posterior activation in people with ASD -> they were activating same region that controls were previously shown to activate more for mentalising tasks CONCLUSION - A non-social task activates a part of the brain that is normally associated with mentalising - Social brain area being used for non-social tasks in ASD ppts - Maybe are re-using this part of the brain for other tasks (like blind people using their visual cortex for other things than vision) - Implies neuroplasiticity = functional specialisation comes about as part of a developmental process

Camille et al 2011

IMPORTANCE People with ventromedial frontal lobe damage are impaired at value-based decision-making AIM - Tested for violations of the General Axiom of Revealed Preference (GARP) GARP - Assumes that individual's choices are consistent with value maximisation if their choices are internally consistent and transitive - If someone picks X over Y, then they should always do so - If they choose X over Y, and Y over Z, then they should choose X over Z METHOD - 9 VMF damage patients and 22 controls - Ppts chose between different pairs of food item bundles - Researchers worked out their number of violations of GARP HYPOTHESIS - An intact ventromedial frontal lobe (VMF) would be necessary for optimum value maximisation and that damage would lead to a greater number of violations RESULTS - Showed that patients with VMF damage were less consistent in their choices CONCLUSION - Suggests that the VMF plays a role in value-based choice - Useful because previous studies have found an increase in activity when making choices, but were unable to show whether the area was necessary Limitations - Isn't able to determine whether the VMF is involved in representing the value of each option, or in making a comparison between values, or both

Volle et al (2011)

Lesion evidence for prospective memory: group studies - Identifying rostral PFC as region - And review of PET/fMRI studies virtually all of which report signal change in that part of the brain

Shallice et al 1991

PATIENT AP: - Male, 23, bilateral frontal open head injury - Given 12 EF and 7 memory tests: all excellent - BUT unable to continue his studies, hold down a job (employers complained of disorganisation) - So current tasks not well measuring the deficits he is showing - Lesion restricted to the anterior PFC - rostral PFC - Maybe more unstructured tasks without clear instructions would be sensitive to his deficit PROPOSAL - Came up with multiple errands test where you have to come up with your own ideas/plans of what to do AIM - Investigated the multi-tasking aspect of executive functions METHOD Multi-tasking: The Multiple Errands Test - Ppts given set of rules and instructions in a shopping task RESULTS - Patients with damage break the rules, don't do things they are supposed to do etc - Examples of Errors: ate purchase (chocolate) before test finished OR Swore loudly

Adolphs et al 1995

PATIENT SM - Damage to the amygdala: left her essentially fearless - Eg. in the pet store, would say she was afraid of snakes but would hold and touch them without any hesitation - Still experienced emotions (like excitement etc), but showed a lack of fear - Bilateral amygdala lesion due to Urbach-Wiethe disease (genetic) METHOD - Asked to rate Eckman faces displaying different types of emotions - Was also asked to draw emotions on faces RESULTS - Specific fear recognition deficit when asked to rate faces on how much they showed an emotion on a scale from 0-5 - Control individuals fear was the same as any other emotion - Also showed a specific fear deficit when asked to draw emotions on faces from memory -> drew a baby for fear - Suggestion that as a child she was frightened by a big dog, so maybe has associated fear with childhood CONCLUSION - Deficits were not because SM is unfamiliar with the concept of fear ( She is also able to describe situations that elicit fear) - Probably experienced fearful situations before damage occurred - Suggests a critical role of the amygdala in fear processing, in line with earlier work with animals

Tsuchiya et al 2009

PREVIOUS WORK - Animal work (eg. using fear conditioning) has suggested a role for the amygdala in the rapid, automatic detection of fear/threat (fast and dirty pathway) AIM - Can Patient SM detect faces covertly? METHOD - Fixation cross - Stimulus presentation (40 msec, subliminal): two faces flashed, a neutral face, and a fearful face at different sides of the screen - Button press at location where emotional (vs. neutral) face occurred RESULTS - SM did not show a sig. diff. to controls, even though she can't overtly identify faces CONCLUSION - But SM throws up some puzzles: she can rapidly detect and process fearful faces, just like controls, despite not having an amygdala - Amygdala not essential for early stages of fear processing

White 2013

The Triple I Hypothesis - Taking Another('s) Perspective on Executive Dysfunction in Autism AIM - Hypothesis proposes that poor performance on EF tasks can be explained by difficulties in forming an implicit understanding of the experimenter's expecations for the task EXPLANATION - Difficulties ASD individuals have with EF tasks may result from difficulties in forming an understanding of experimenter's implicit expectations - Impairment in Inferring Implicit Information (Triple I) - Leading to idiosyncratic or egocentric pattern of performance - Especially in open-ended tasks, because thy allow the individual to impose their own demands on the task situation - Or they involve arbitrary rules, which ASD individuals ignore (Russell, 1997), because they are unaware of the social expectation that they should follow such rules CONCLUSION - Explains inconsistency in results of EF tasks because sometimes the child's demands align with the experimenter's and sometimes they don't - Might even explain why there is decreased activity in FFA when ASD ppts just shown pictures of faces, vs. when they are told explicitly to look at the eyes of the faces (and activity increases) - Basic face processing thus appears to be intact, but isn't shown because the ppt doesn't understand the experimenter's implicit demands