21 Pages

Course Code
John Vervaeke

This preview shows pages 1,2,3,4. Sign up to view the full 21 pages of the document.
PSY370 FINAL EXAM STUDY GUIDE (Post-Midterm) PART 1 INSIGHT Recent Insight & Neuroscience Research October 27 Weisberg comes back and makes a methodological critique, saying insight research has conceptual confusions about: 1. Subjective status of insight problems – how can we determine whether a “problem” actually exists? What is a problem for Trevor may not be a problem for me (the concept isn‟t stable or objective) a. Bowden & Jung-Beeman 2. Homogeneity of CLASS of insight problems: maybe there are hybrid problems, some require search- inference as well as insight processes. If so, our performance on “purebred” insight problems are not generalizable a. Gilhooly & Murphy  how can we determine objectively what constitutes an insight problem? BOWDEN & JUNG-BEEMAN 2003 Aha! Insight experience correlates w/ solution activation in the RH.  told subjects what is MEANT by “insight” & “non-insight” & the associated experiences  subjects given a bunch of non-labeled problems, told to REPORT when they are having an insight experience.  Results: when patients were working on problems that they reported to be associated with insight, they showed a quick hemispheric lateralization to the RH. Across subjects, the same problems were associated with the same reports and the same type of brain activity (lateralization) o Specifically, in the right anterior supramarginal temporal gyrus, EEG showed increased activity almost immediately before solution  Correlates with F.O.W. data! (Metcalfe‟s feeling-of-warmth theory) Their results were consistent & theoretically meaningful The Theoretical Meaningfulness of Hemispheric Lateralization To get a solution to an insight problem, we make a sudden shift to a landscape/gestalty view of the environment (RH), and then a quick shift BACK (LH) - probably verifying our solution. Explanation: When 2 organisms compete, the one that wins is able to pick up on small variations & details & can make smaller changes in his behavior ----- fine-grained & sequential processing ~ occurs in LH Gestalty big-picture processing and response required for NOVELTY & multiple simultaneous constraint problems (such as being attacked by a tiger)~ occurs in RH navon letters (bigH made up of Ss) –  damaged RH will see only sssss  damaged LH will see only H  WE see both – approach a dynamical equilibrium - both seem to be pulling on each other Snyder & Chi THINKING CAP transcranial magnetic stimulation  impose lateralization shifts  improve insight probsolv! ****When somebody can use previous results & theories and CAUSE/ INTERVENE in phenomena, it‟s strong evidence for UNDERSTANDING of a phenomenon**** ** insight problem-solving lateralization reflects a shift in processing STYLE, not content. BOWDEN & JUNG-BEEMAN point out that LANGUAGE PROCESSING shows this shift as well, triggering spikes in RH activation  garden path sentences: the horse raced past the barn fell  metaphor: shifting levels of construal shows that “VERBAL OVERSHADOWING” is a misnomer – the problem isn‟t language in and of itself, it‟s that a specific type of concurrent verbalization causes overshadowing. The STYLE & MANNER OF PROCESSING is what‟s important LANGUAGE THAT OVERSHADOWS: the kind that results in LH routine sequential processing LANGUAGE THAT FACILITATES INSIGHT: the kind that results in shifting levels of construal, triggers spike in RH big-picture activation. Gilhooly & Murphy 2005 Differentiating insight from non-insight problems  if restructuring is necessary ~ it‟s an insight problem  if it‟s not necessary ~ non-insight problem  if it‟s helpful but not necessary ~ hybrid problem A) cluster analysis  look for a positive manifold (things are mutually predictive & highly correlated) in insight problems – assumes that different problems share core processing features o FOUND IT! Insight probs clustered strongly together, non-insight too, but less so. o Empirical results correlate w/ a priori taxonomy for insight vs. non-insight problems o  our intuition of insight probsolv seems to reflect reality B) individual differences methodology  see WHICH abilities are predictive of insight & non-insight problem solving o cognitive flexibility is important for insight (requires attention shifting, inhibition of no- longer relevant info, working memory)  facilitates dynamical processing (LH – RH shifting) TASKS:  figure fluency task – make as many shapes as possible from a dot pattern (requires gestalt  detail switcherooing)  alternative use task - how many things can you use this object for? Weisberg comes back again w/ Fleck … ☆IS THIS IMPORTANT???☆ Trying to pay attention to methodological work on verbal protocols Lotta people have been noticing interference effects, retooling how we use concurrent verbalization REVIEW – re-design their constructions on how to engage in concurrent verbalization Arguing against SCHOOLER‟s thing that concurrent verbalization impairs insight problem solving Facial recognition & verbal overshadowing  see faces, later describing them impairs later recognition OVERVIEW OF CONCURRENT VERBALIZATION RESEARCH Macrae & Lois show that language isn‟t necessary for overshadowing in FACIAL RECOGNITION  see faces  shown something like NAVON LETTERS a) driven to FEATURAL level (LH; sss) b) driven to GESTALT level (RH; H)  facial recognition task o a) featural shows overshadowing o b) gestalt outperforms controls!!  facilitation! Schooler says it ain‟t verbal overshadowing, it‟s a Transfer Inappropriate Processing Shift *** Suggests that we could facilitate probsolv by changing the intervening task***  FINGER 2002  Participants presented with pictures of faces  Participants describe faces aloud  Intermediate task o group a: do maze or listen music (MSC, pattern detection& tracking)  no impairment o group b: do another verbal task  overshadowing  Facial recognition task conclusions: group a‟s second task shifted them BACK to gestalt level Meisner et al. 2001 ☆ More extensive & constrained/demanding verbalization task  more overshadowing (inapprop processing shift) (extensive = asked to elaborate, be more thorough/detailed) Allowed to be loose & associational  used language that doesn‟t trigger LH activation  no overshadow Why? – this requires consistency, making judgements about the coherence of your descriptions – search inference processes!!! CONSTRUAL LEVEL THEORY Altering construal – shifting the SCALE of your attention – seems crucial in problem-solving Abstract Concrete …People can manipulate it in many ways FORSTER & Liberman 2004 Temporal Construal Effects on Abstract and Concrete Thinking: Consequences for Insight and Creative Cognition Temporal Imagination – while solving a problem, either imagine you‟re solving it NOW, or 1 YEAR from now Imagining solution one year from now ENHANCES insight! Something “far” – physically OR temporally – you can‟t “see” details  see abstract big picture  HUNT & CARROLL 2008 Verbal overshadowing effect: how temporal perspective may exacerbate or alleviate the processing shift Almost exact replication of Forster & Liberman, more evidence that altering construal is CRITICAL in insight probsolv. VERVAEKE: suggests that altering construal requires opponent processing: where 2 functions perform opposite & conflicting tasks but are necessarily integrated – like the parasympathetic & sympathetic nervous system – and from this dynamical system, a “design” naturally emerges Featural level:  breaking up inappropriate fixation, allows detail|  overshadowing (autistic symptoms?) Gestalt level:  allows insight, big picture  causes fixation, (schizophrenic symptoms?) ^ complementary set of strengths & weaknesses It would be great design if the brain does opponent processing between the two  dynamical, self- organizing system, probably running on a neural network of pattern tracking & construal - seems likely because attention shifting has to be ORGANIZED, and ISN‟T done DELIBERATELY schiz & autism stuff  theory of salience dysregulation, weak central coherence theory (autism) DEYOUNG, Flanders & Peterson 2008 Cognitive Abilities in Insight Problem Solving: An Individual Differences Model. Trying to figure out the mechanisms of insight problem solving – say it‟s RESTRUCTURING, which requires Notes INDIVIDUAL DIFFERENCES between various measures  convergent thinking: logic, reasoning, good for well-defined problems  breaking frame: ability to overcome context-induced fixation  divergent thinking: pattern recognition, intuition (like for blurry picture tasks), includes cognitive flexibility each was predictably related to insight probsolv, but only the last two were UNIQUELY predictive – and the only SUBcategory of divergent thinking that was predictive was !!!cognitive flexibility!!! PROBLEM: didn‟t talk much about the machinery of/alternatives to breaking frame. POLYANI - transparency  opacity shift  learning-to-learn (it‟s like if your glasses became opaque – you can SEE the process by which you‟re seeing) learning = noticing a pattern in the world learning-to-learn = noticing a pattern in the learning medium cognitive message to cognitive medium thought experiment on attentional scaling: hit a pen against a brush, to notice the shape of the brush. THEN notice the vibrations of the pen itself  brush dissolves perceptually. ☆ tied to mindfulness!! APTER used same terminology. It‟s also a gestalt  featural shift – noticing things that COMPOSE my noticings Opacity  transparency shift  predictive of INSIGHT! (Probably transopac comes first, though) 2D model of breaking frame: Gestalt  Featural, Transparency  Opacity LH 2D model of making frame? Opacity  Transparency, Featural  Gestalt  RH (result of breaking?) BREAKING + MAKING = ATTENTIONAL SCALING Knoblich & Ohlsson et al 1999 Constraint Relaxation and Chunk Decomposition in Insight Problem Solving Matchstick stuff. Constraint: salience/relevance being projected onto a problem to restrict search space The probability of constraint relaxation is inversely proportional to its scope: measure of hierarchy of constraint higher order constraint  more accommodation required  less chance of relaxing constraint. More assimilatory the change/lower order  less reconfiguring (of mental set) required  easier to relax constraints Piagetian assimilation: incorporate new information to your structure vs. accommodation: change your structure to fit in new information. Chunk: pattern of co-relevance among data. No explanation of how we do it. Probability of chunk decomposition is inversely proportional with its tightness (degree to which you can‟t break it up – relative to level of constraint??) CIRCULAR DEFINITION!! KERSHAW & OHLSON 2004 Multiple Causes of Difficulty in Insight: 9-Dot Problem Constraint Factors  perceptual figural integrity etc… -- pattern factors is a better term  processing formulation factors  knowledge transfer factors All tightly interdependent & mutually reinforcing Experiment: noted that facilitation effects in research were small, so they removed a varied # of difficulties in 9-dot problem The more CONSTRAINT FACTORS are addressed, the more solution of insight problems are FACILITATED but not completely. All factors reinforce each other, so giving 3 individual pieces of info doesn‟t necessarily allow you to coordinate them into a useful whole – individual skills ≠ expertise  “DYNAMICAL SYSTEM OF CONSTRAINTS” requires a system of procedures, a meta-skill, in addition BOWDEN et al 2005 New approaches to demystifying insight. ☆☆☆☆☆☆ Recent Insight Theories Nov 3 NEURAL NETWORK THEORY HOW THEY LEARN: input  � output  take output value, compare to target value (correct response)  error value  change connections between nodes a wee bit (BACK PROPAGATION OF ERROR) Unsupervised Version: building target values into the brain – Evolutionary Module – coarse, general, around for a LONG TIME, necessary but not suffictient  Internalization: a & b boxes Phase 1: projection problem Phase 2: i) A stands IN for world, B improves its skill at solving projection problem (learning to learn) ii) B meta-models, A models, B supervises modeling technique Geoffrey Hinton says there are many stages. Calls 1 Wake, 2 Sleep/Dreaming In reality there‟s prolly hierarchical convergence, it ain‟t just linear or 2 things. MANY As (WM, vision, somatosensory stuff, etc…, sharing the same metamodellers)  B can recognize similarities (things that CONVERGE) between As in a highly abstract procedural way  mapping abstract procedural problems from one procedure loop to another … = INSIGHT!!! Vandervert et al 2007 How Working Memory & Cerebellum Collaborate to Produce Creativity & Innovation Evidence in brain for this typa mapping! Parts of cortex & cerebellum. Cerebellum = B-A rel.ship center Numbers allow convergent internalization- we can hear, touch, feel, see “three/3/ | | |” SYNAESTHESIA: mapping things onto completely different modalities Ramachandran Buba & Kiki – naming shapes is totally making sense. Buba looks bloopy and Kiki looks sharp and spiky Liu & Kennedy 1993: SQUARE & CIRCLE w/ QUALITIES People had a ginormous pattern of agreement that circle was assoc w/ soft happy mother bright alive summer while square was assoc w/ hard sad father dark dead winter DID NOT TRANSFER TO METAPHOR DEBARTOLO et al 2009 Cerebellar involvement in cognitive flexibility Cerebellum impairment decreases learning-to-learn in lesioned rats, but NOT learning! Other cerebellum research:  cross-modular transfer affords internal models, & switching thereof, having to do w. HIGHER-ORDER functions like mindsight  transfer in learning & learning to learn The cerebellum does INTERNALIZATION & COGNITIVE FLEXIBILITY. But not exclusively - Hippocampus maybe also internalizes. th November 10 Murry & Byrne - ATTENTION & INSIGHT ability to SHIFT ATTENTION is predictive of insight Selective attention ALONE isn‟t, ability to FIX attention isn‟t Synesthesia + + creativity, machinery that explains insight explains synesthesia MONTAGUE a neuroscientist Computer – runs hot & fast (uses up LOTS of energy real rapidly) People – run warm & slow (never get direct feed of power – food) batteries could kill you/you could be somebody else‟s battery. You have to worry about logistics as well as logic  humans pursue efficiency PROBLEM: we have 2 opposing goals – - coordination & organization (requires LOTS of cross-talk in the brain) - fewest connections possible (wire is metabolically costly – gotta REDUCE cross-talk)  MUTUAL MODELLING: things form internal models of each other, consult it  highly coordinate behavior. Update models by periodically communicating. Montague says parts of our brain do this cross-talk: modules consulting each others‟ models RIMM- recursive internalized mutual modeling: self-organizing systems, unlike mathem. knowledge SELF-ORGANIZATION OF DYNAMICAL SYSTEMS Recursive, FEEDBACK – the output of the system can be fed back  input. How order arises – interaction between individual members of a system an alternative explanation to randomness & intelligent design DYNAMICAL SYSTEMS THEORY: Interaction: relation between events that CHANGES ACTUALITY ex: you push a chair Conditions: constraints that ALTER POSSIBILITY ex: there‟s no obstacle in front of it Evolution: morphology (genetics etc)  behav.  INTERACTION W/ENVIRONMENT & environment  shapes conditions (in environment)   natural selection  MORPHOLOGY AGAIN causal interactions of component parts  overall component structure  sets conditions/constraints for how parts can interact STEPHEN & DIXON 2009 The Self-Organization of Insight: Entropy and Power Laws in Problem Solving About dynamical systems-theory Component-dominant = top-down processing Interactional dominant = bottom-up processing Phase Transition: one structure quickly reorganizing into a different form after criticality STEPHEN & DIXON 2009 PHASE TRANSITION: Experiment: given pictures of buncha interlocking gears in different configurations, hafta figure out what will happen at Z if you move A in a certain direction a) force tracing: trace the “force” through the gears like in a line (not transferable) b) alternation: even # of gears – clockwise, odd # - counter-clockwise – (transferable) 1) system of mutually interacting & constraining parts within a whole 2) entropy overwhelms the system‟s constraints (input to system can‟t be processed stably by the system) 3) critical instability  change  affords NEW STRUCTURE (opposite of entropy) 4) return to 1), with a new, more COMPLEX system, that can handle the amount of entropy that previously overwhelmed it ☆ this is what happens when we do breaking & making frame!! Experiment 2: “noise” added (distracting irrelevant info ~ entropy) Adding some entropy increases peoples‟ ability to solve insight problems (works w/ returning act hypothesis) Making frame: ability to incorporate higher-order relations that were originally noise to the system Lower-level: short, detailed, local relations High-level: globaller. Right before insight solution, there‟s a massive increase in entropy & disorganization Correlates w/ FOW IMPORTANT PROPERTIES OF DYNAMICAL SYSTEMS 1) MULTISTABILITY Pile of sand  loss of success & increase in entropy  broader base  new stable structure ….self-organizing criticality: increase in entropy affords new stability. Constraints relaxed, chunks decomposed 2) COMPLEXIFICATION Cause complex things have EMERGENT functions ...Due to alternations of integration & differentiation between parts… works ok w/ sand, better w. EVOLUTION the brain is a machine that can improve itself local timescale: micro-event (grains of sand bumping) developmental: big-picture happenings (pile) SCHILLING 2005 A “Small-World” Network Model of Cognitive Insight Network Theory, not neural Small world networks: regular (equidistal connections) network + couple of long-distance connections Small loss of order  huge increase in speed of moving between points - lower path length What if your brains are constantly trying to build these Coefficient of connectivity: how well 2 dots picked at random are connected. High order, mutual predictability. Small world networks = most efficient These networks are deeply connected to SOC (self-organizing criticality of opponent processing between integration & differentiation) & to complexifications. SOC tends to CREATE SMNs YOUR BRAIN ACTUALLY WORKS THIS WAY! SOC  SMN groups of neurons sync together (integrate), avalanche,  asynchronous (differentiate)  resync (opponent processing) This maps on to making & breaking frame. We can correlate sync/async variations w/ ability to solve problems…. & predict GENERAL INTELLIGENCE STANKAR et al. 2006 Individual differences in ability to sync & maybe desync predict variance in G THATCHER et al. 2009 avalanche (☆in what capacity??) & then come back has HIGH correlation w/ individual differences in general intelligence Ability to break & make frame is indicative of variance within general intelligence Flexibility is also correlated with this Specified idea of reconstrual from Gestaltists can be found in so much science Insight: dynamical system  reconstrual  opponent processing NOT COMPUTATIONAL We have passed the action/perception distinction  ENACTION Mindfulness, Intuition, Creativity & Flow Nov 17 Reason: alteration of belief through inference ☆CREATIVITY☆ grandfather definition: human capacity for analogy allows fruitful transfer from previous situations DEIRDICH GENTNER: “STRUCTURE MAPPING THEORY” Problem? “structure” = about content, propositional, but analogy is PROCEDURAL Looks @ the solar system model of the atom – it‟s WRONG (nucleus & orbiting stuff) misses quantum mechanics … but it‟s good scientific creativity – analogous to orbiting of planets & sun - not completely the same - only SOME of the propositional content is transferred Predicates:  attribute – takes 1 instantiation ex: “large (x)”  relation – takes 2 or more ex: “collide (x,y)” - first-order: take OBJECTS as instantiation - second-order: take PREDICATES as instantiation. Ex: CAUSE [collide(truck, car), strike(woman)]  Hierarchy of Constraints on Transfer for Analogy 1) prefer RELATIONS to ndtributes 2) in relations, go for 2 -order predicates first so, for the orbiting metaphor, it‟s EQ[Attraction (x, y), Orbiting (y, x)] this sums up the “important” parts of solar system AND atom… higher order structure of relations SYSTEMATICITY: property of organization such that the more the above can be done, the more stuff is similar Self-complexifying systems seek this out Grabbing hierarchy of the structural relations of one system and applying this hierarchy to another system Systematicity + + Complexification (simultaneously integrating and differentiation) PROBLEMS 1) ATTRIBUTES & RELATIONS ARE ILL-DEFINED – logic doesn‟t distinguish between the two. Similar to GPS problem. Plus logic doesn‟t distinguish between the two  combsplosh of # ways to represent information. Doesn‟t address how we formulate problem Gentner replies Interested in how things are psychologically represented, not logical, Suggests exhaustive algorithmic search – Looks like a computational theory of insight problem-solving 3) What‟s the proper level of specificity? Problem of level of construal 4) Insight machinery ++ analogy machinery 5) chicken & egg problem: you need to structure the unstructured data before even searching for the structure – like T/-\E C/-\T Analogy can‟t explain insight because it fundamentally presupposes insight JOHNSON-LAIRD: little transfer is driven by pure structure. Humans aren‟t good at reasoning w/ structure IS CREATIVITY DIFFERENT FROM INSIGHT? BOWDEN 1990 says NO!!! creativity == insight Creative: solve a problem you haven‟t been able to before Maybe creativity‟s goal = problem FINDING (&play) , insight‟s goal is problem SOLVING (work) Henle @ creativity symposium 1961 detachment: crucial for creativity construal is important for creativity CRUTCHFIELD 1962 - intrinsic (own sake) vs. extrinsic (goal external to task) motivation Says Extrinsic motivation prevents detachment, because your work becomes transparent Reduces shift towards learning-to-learn AMABILE, 1993 Motivational Synergy: Toward new conceptualizations of intrinsic and extrinsic motivation in the workplace Intrinsic motivation is conducive to creativity & insight, Extrinsic is DETRIMENTAL CONCEPTUAL CONCERNS:  hard to distinguish types of motivation ppl have. Ex: achievement  some forms of extrinsic motivation can improve creativity  COLLINS & AMABILE 1999 looked at which features of extrinsic motivation were detrimental to creativity - dividing attention between current task & future goal  decreased focus STERNBERG & LUBORT Synergistic & Non-Synergistic EXTRINSIC Motivation (talking about flow as synergistic – feeds back & helps creativity, working in concert w/ intrinsic motivation, cause it gets you to look outside the task to a source of feedback on how you‟re doing on the task  FACILITATE (idea of motivational work cycle) suggest task-focusing vs. goal-focusing APTER 1984 Reversal Theory and Personality: A Review THEORY OF META-MOTIVATION Arousal: physiological Motivation: interpretation/framing of arousal (negative/positive valence? Approach vs. avoid) DRIVE REDUCTION THEORY- FREUD: classic model of motivation (false) HEBB’S THEORY OF OPTIMAL AROUSAL (parabolic graph) APTER 1989 pointed out 4 problems with Hebb‟s 1) predicts that relaxing = higher level of arousal than excitement! 2) excitement peaks @ mild-medium, relaxation‟s minimum is mild-medium... 3) anxiety & boredom aren‟t only found at extremes of arousal 4) predicts a sequence. To get to anxiety from boredom, have to be excited first?! Level of arousal ≠ level of Affect  APTER’S ALTERNATIVE: Bistable Model of Arousal multi-stable system – dynamical systems language Intrinsic vs. extrinsic motivation, Paratelic: cognitive play, creativity – excitement/boredom – intrinsic (^ arousal = ^ involvement) Telic: cognitive work, insight – anxious/relaxed – extrinsic (activity for sake of GOAL) (operationalization of “work” & “play”) Takes into account that same levels of arousal can have different hedonic states  different feelings Reconstrual of arousal between meta-motivational modes What‟s the difference between the two? FRAMING!!!!! ☆☆ ASPECT SHIFT ON AROUSAL☆☆  how can we get people to FLIP between telic & paratelic? (restructuring our interpretation of our own levels of arousal) like in roller-coasters! motivation to look for interesting problems - non-specific problem finding: If I can get a certain formulation that‟s MULTI-APT, I can get at solving other problems! The formulation is like a nexus that can be applied to many different problems  cog. complexification!! Differentiate between problems but interconnect them with a single problem-finding formula. Cognitive synergy: note
More Less
Unlock Document

Only pages 1,2,3,4 are available for preview. Some parts have been intentionally blurred.

Unlock Document
You're Reading a Preview

Unlock to view full version

Unlock Document

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.