ההשפעה של מוטיבציה על התנהגות

  • Published on
    08-Feb-2016

  • View
    66

  • Download
    0

Embed Size (px)

DESCRIPTION

. ?. Pavlov ( ) Thorndike ( ) Skinner () - PowerPoint PPT Presentation

Transcript

<ul><li><p> ?Pavlov ( )Thorndike ( )Skinner ()</p><p> () , ...</p><p>"The fact is that the study of psychological processes controlling simple goal-directed instrumental actions by primary motivational states has been neglected since Hull and Tolman (40s) (Dickinson+Balleine 1994)</p></li><li><p> :Associative learning is not a simple, unitary phenomenon " , . </p></li><li><p> ? ( )Primary motivation / ( )Incentive motivation " (" "/" " )</p><p> (directing) (energizing)</p></li><li><p> (UR) : , , -US -CR ( " CS -US) ( " S-R)(Rescorla, Holland, Colwill)</p></li><li><p> Cardinal et al. (2002)</p></li><li><p> ? primary motivational shift ( ) ( ) !Sodium appetite ( ) </p><p> post-training shift in outcome valueConditioned aversion (LiCl) Specific satiety ( )</p><p> , ( ). .</p></li><li><p>, ? </p></li><li><p> () goal directed action " - (A-O) habitual behavior " A-O " S-R</p><p> ? " (contingency degradation) " (LiCl specific satiety) ( ) </p></li><li><p>1. : irrelevant incentive , ( Y-maze, Adams )</p><p> " - ... " " (incentive learning) ( ) , Dickinson+Balleine - ( ) : Rescorla LiCl, Dickinson -incentive learning.</p></li><li><p>2. - ? overtrainingAdams , \, Adams + Dickinson VI -VR ( LiCl ); Dickinson .</p><p> Rescorla+Colwill -R-O, ( -choice extinction test)</p><p> ( ) ( ).</p></li><li><p> ? ... Balleine+Dickinson , . , ( ). " - ( , , ()).</p></li><li><p> ..Balleine, Dickinson et al. 1995 - -instrumental chain ( ; ) magazine approach ? ( )?</p></li><li><p>... -Pavlovian Instrumental Transfer (PIT) - general motivating effect CS , outcome ( ) "conditioned motivation. (same&gt;different) -CS -US , -US -CS . : , . " (DA antagonists -PIT, , -Incentive learning ).</p></li><li><p> - ?: . Amygdala:BLA -CS -US CeN controller , S-R Nucleus Accumbens: -R-O " PIT. (delayed)Cortex:Prelimbic R-OInsular Cortex " Orbitofrontal Cortex -BLA, (instrumental choice behavior)Anterior Cingulate .</p></li><li><p> ( ) ? -scope , ( ) . </p></li><li><p> Corbit+Balleine 2000 - , - (). , , . " R-O, ( Dickinson).: -S-R -postconditioning outcome devaluation. 1 specific satiety , , choice extinction test (VR) -test</p></li><li><p> Corbit+Balleine 2000 ? 2 contingency degradation -degradation ( )-test -nondegraded ...</p></li><li><p> Corbit+Balleine 2000: ( ), R-O . -context ( ) predictive validity ( ). R-O. " ( ... ?)</p></li><li><p> Knowlton, Mangels and Squire 1996 "" (+ ) ( ) ( -basal ganglia) ( ), ( ) Double dissociation </p><p>Pavlovian associations:CS-CR demonstrated convincingly in second order conditioning (see below)US specificity (CS-US association) demonstrated by sensory preconditioningThe direct link of the CS to affect is demonstrated in trans-reinforcer blocking (example the presence of an aversive CS (paired with shock) can block the conditioning of a CS signaling the omission of an appetitive reward the two reinforcers share nothing except their aversiveness, hence the blocking effect must depend on an association between the CS and affect). </p><p>Pavlovian conditioning gives rise to at least two types of associations. One is between a representation of the CS and a representation of the US. This gives rise to stimulus substitution effects (in which the animal treats the CS as if it were the US), and a class of CRs which reflect the specific sensory-perceptual properties of the US, and have been termed ``consummatory responses'', such as salivating to food, licking to water, eye-blinking to a puff of air, etc. A second type of association is formed between the CS and a generic appetitive or aversive motivational system (depending on the class of USs used). This direct association does not depend on the specific sensory properties of the US, and allows the CS to activate non-US-specific preparatory CRs, such as approach or withdrawal (termed ``preparatory responses''). The motivational system can also be activated by the CS indirectly, via a link between the US representation (which is associated with the CS) and the appropriate motivational system. Dickinson and Balleine (2002) hypothesize that the indirect activation of the motivational system by the US representation is gated by the motivational importance of the US. Thus, for instance, the strength of a consummatory response to a water US is monotonically related to the degree of thirst. Thus, although Pavlovian CRs are not controlled by the contingency between the conditioned response and the US, their performance can be sensitive to the level of motivation for the US. Furthermore, associations can also be formed directly between the CS and the CR, as has been shown in secondary conditioning paradigms. The direct CS-CR association can be demonstrated by pairing the first order CS with a new US eliciting a new CR, which does not affect the prior conditioning of the second order CS, which continues to elicit the original CR. In inhibitory conditioning, the existence of CS-CS associations has also been implied (Rescorla, 1982) by the incomplete transfer of inhibition to a new CS</p><p> Affective/hedonic (Garcia), (incentive value)CS , Hebb 1949 .</p><p>The primary candidate for the DA-dependent pathway presumablymediating Pavlovian incentives (i.e., driving Pavlovian behavior), isthe nucleus accumbens (NAC, also referred to as ventral striatum,comprised of core and shell regions). \citeA{RobbinsEveritt1996}stress the role of the NAC in mediating Pavlovian motivation forappetitive rewards, especially in preparatory and not consummatoryresponses, as extracellular DA concentration in the NAC peaks at leverpress and then goes down as the animal retrieves the reward(\citeNP{KiyatinGratton1994}, using chronoamperometric techniques, andsee \citeNP{PhillipsEtAl2003}, for more recent fast scan voltammetrymeasures with sub-second resolution). Furthermore, the effects ofPavlovian stimuli on instrumental behavior in PIT depend crucially onDA-dependent interactions between the NAC and its limbic afferents (inparticular the basolateral nuclei of the amygdala (BLA)).</p><p>In a recent study, \citeA{CorbitEtAl2001} elaborated the anatomicaldistinction between the two pathways by demonstrated a doubledissociation between NAC core-mediated, DA-independent, effects of amotivational shift treatment, and NAC shell-mediated DA-dependenteffects of PIT. Whereas NAC core lesions impaired instrumentallearning, and disrupted the effect of specific satiety devaluation,shell lesioned and sham controls exhibited a normal devaluationeffect. However, in PIT the sham and core lesioned groups showednormal transfer, while the shell lesioned group did not, demonstratingthat the NAC shell mediated the effects of PIT. Lesions to theCentral nucleus of the Amygdala, but not to the BLA, have also beenshown to abolish PIT \cite{HallEtAl2001}.</p><p>The BLA, in turn, is implicated in the DA-independent instrumentalincentive pathway, as the structure crucial for integration of outcomevalue into the A-O association, in order to guide performance. In astudy of the role of the BLA in instrumental learning,\citeA{BalleineEtAl2003} showed that BLA lesions did not affectinstrumental learning curves, or discrimination between two actionsleading to two outcomes, but disrupted the effects of specific satietytreatment, similar to the results of NAC core lesions. Furtherexperimentation showed that this effect was due to the inability ofthe lesioned animals to associate or represent the sensory or hedonicproperties of the different outcomes\cite{BalleineEtAl2003,KillcrossBlundell2002}.%The authors concluded%that BLA lesions impair the ability to discriminate between two%outcomes based on their sensory properties, and thus attenuate the%ability of the outcomes to control behavior differentially. However,%the outcomes are still able to engage a general reinforcement%mechanism, by which instrumental learning can proceed, possibly%through an S-R controlled process, which is sensitive only to the%general reinforcing impact of instrumental outcomes. Thus they argue%for a role of the BLA in ``liking'', and in the evaluation of the%sensory-specific hedonic value of an outcome. This places the BLA%(with its strong connections to the NAC core) as a fundamental%component in instrumental learning, through which outcome value is%integrated into the A-O association, in order to guide performance.Similarly, the dorsolateral prefrontal cortex (and its homologue, theprelimbic area of the prefrontal cortex, in rats) has been implicatedin response selection and planning using A-O knowledge. Lesions inthe prelimbic area produced a general decrement in both responsesafter the instrumental contingency of only one of the responses isdegraded, and resulted in insensitivity to specific satietymoticational shift, implying the use of S-R knowledge to controlbehavior in the absence of ability to utilise or encode A-Oassociations \cite{BalleineDickinson1998}.</p><p>%\citeA{BalleineDickinson1998} further showed that the insular cortex,%which has been implicated in gustatory processing, may be a site for%memory of incentive value of food, based on taste. Lesions of the%insular cortex were shown to have no effect on responding after%degradation of contingencies, but to abolish the effects of specific%satiety by prefeeding, as animals continue to choose the action%leading to the prefed outcome and the action leading to the non-sated%outcome equally. However, in this case, when the choice test was not%conducted in extinction, the animals were able to show a preference%for the non-devalued food. This implies that the insular cortex is%necessary for retaining information about an unavailable outcome's%value, but not for assigning the value to an available reward.</p></li></ul>

Recommended

View more >