Is Our Self Nothing but Reward.pdf

7/27/2019 Is Our Self Nothing but Reward.pdf

1/7

REVIEW

Is Our Self Nothing but Reward?Georg Northoff and Dave J. Hayes

Neuroscience has increasingly explored the neural mechanisms underlying our sense of self. Recent studies have demonstrated the recruit-

ment of regions like the ventral tegmental area, ventromedial prefrontal cortex, and the ventral striatum to self-specific stimuliregions

typically associatedwith reward-related processing. This raisesthe question of whether there is a relationshipbetween self andreward and,

if so, how these different fields can be linked. Three relationship models that aim to explore the relationship between self and reward arediscussed here: integration, segregation, and parallel processing. Their pros and cons are reviewed in light of the most recent findings. The

conclusion is that both the fields of self and reward may benefit from increased interaction. This interaction may help to fill in some of the

missing pieces regarding reward-related processing, as well as illuminate how brain function can bring forward the philosophical concept

and psychological reality of self.

Key Words: Animal, human, models, psychiatric disorders, reward,

self, translational

Though the self is originally a philosophical concept, it hasbecome the subject of intense neuroscientific investigation,especially in neuroimaging (13). This research has revealed

that several brain regionscollectively referred to as cortical mid-

line structureslike the ventromedial prefrontal cortex (VMPFC),dorsomedial prefrontal cortex (DMPFC), anterior cingulate cortex,and the posterior cingulate cortex, are recruited during exposure toself-specific stimuli when compared with nonself-specific ones.(Self-specific stimuli are considered to be those that an organism

identifies as highly personally relevant.) However, there has beensome doubt about the strict association of the cortical midlinestructures with self-specific stimuli because some familiar and un-familiar nonself-specific stimuli also induce activity changes inthese regions (4,5).

Recent studies on self and reward have demonstrated that self-specific stimuli induce neural activity changes in regions recruitedduring reward, e.g., VMPFC, ventral striatum (VS), and ventral teg-mental area (VTA) (6 9). Theapparent overlap in neural activations

between self-specificity andrewardraises several questionsregard-ing their relationship. For instance, is self-specificity nothing but

reward? In other words,does self-specificity really amount to thosethings that have high degrees of value for us? Do we need toconsider that the existence of the self is an illusion as argued bysome philosophers (10)?

Despite an increased understanding of the variety of functionssubsumed under the concept of reward (11,12), several issues re-

main open in this field. Forinstance, Montague (13) haspointed outthe relevance of linking incoming exteroceptive stimuli to thebodys interoceptive stimuli by assigning value, and thus reward, totheformer. This maybe crucial because theexteroceptive stimulusrewarding features may also depend on the organisms actualstates and hence its interoceptive stimuli. Self-specificity may be

constituted by linking interoceptive and exteroceptive stimuli (6),which mayallow a certain continuity over longertime intervals(i.e.,self-continuity; see Ersner-Hershfield et al. [7]). Hence, some of the

open questions in the field of rewardmay interface with issuesandproblems raised in the field of self.

The aim of this article is to explore the interface between rewardand self by discussing different models of their potential relation-ship.Basedon theactual data,we propose threedifferentmodelsofthe relationship between self and reward to structure the abun-dance of current empirical findings in both domains: overlap be-tween self and reward (i.e., the integration model), nonoverlapbetween both (i.e., the segregation model), and self as parallelprocessing at some level (e.g., higher or lower order processes)along a rewardcontinuum(i.e., the parallel processing model) (Fig-ure1). Evidence acrossstudies supporting andrefutingeach modelwill be offered and discussed. Along with evidence in humans,some nonhuman animal studies that have implicated the involve-ment of similar brain regions in reward-related, and potentiallyself-specific, processing will also be considered. It is also importantto note that although the translational evidence discussed maysupport one model or another, the original investigators of eachstudyunder consideration maynot agree with such anapproach; assuch, it is acknowledged that other possible interpretations mayexist. Finally, the potential impact of considering the relationshipbetween self and reward on psychiatricdisorders, suchas addictionand depression, is indicated (see Table 1 for summary and Supple-ment 1 for a more in-depth discussion).

Reward-Self Relationship Models

Before discussing the reward-self relationship models, it is im-portant to note that the terms reward and self are used here in avery basic sense (compared with the various uses within their re-spectivefields; forexamplesin thefieldof reward, see[1417]).Onecore component of reward is value assignment, meaning that anexternal stimulus is assigned specific relevance or importance forthe organism (18,19). Similarly, although there are numerous as-pects of the self (e.g., self-recognition, self-consciousness, self-awareness),the use in thepresentworkrefers largely to thefactthatsome stimuli areself-specific. This basic or core aspect of self canbeconsidered the primal characterization of the organism in relationto the environment, and thus, this concept may be shared amonghumans and other animals (20).

First Model: Integration of Self and RewardThe integration model of self and reward claims that our self is

almost entirely reward; that self-specific stimuli are really thosethathave a high level of value assignment for us. Self-specificity maythus be constituted and distinguished from nonself-specificity bythe value of the stimuli. This may hold not only for external stimulifrom the environment but also for internal (i.e., interoceptive/bodily andcognitive/mental) stimuli. These too must be assigneda

From the Mind, Brain Imaging and Neuroethics Unit, Institute of Mental

Health Research,Royal Ottawa Health Care Group, University of Ottawa,

Ottawa, Ontario, Canada.

Authors GN and DJH contributed equally to this work.

Address correspondence to Georg Northoff, M.D., Ph.D., University of Ot-

tawa, Institute of Mental Health Research, 1145 Carling Avenue, Room

6467, Ottawa,ON K1Z7K4, Canada; E-mail: [email protected].

Received Sep 23, 2010; revised Nov 24, 2010; accepted Dec 14, 2010.

BIOL PSYCHIATRY 2011;xx:xxx0006-3223/$36.00doi:10.1016/j.biopsych.2010.12.014 2011 Society of Biological Psychiatry
mailto:[email protected]:[email protected]:[email protected]


2/7

value to be self-specific and distinguished from nonself-specificones. Hence, value assignment to stimuli is not only at the core ofrewardbut mayalso be crucial in constitutingself-specificity anditsdistinction from nonself-specificity.

Support for the integration model comes from human imaging

studiesbyPhan etal. (8) anddeGrecketal. (9).Phan etal. (8) showedthat during the viewing of emotional pictures, neural activitychanges in the VS and the VMPFC are linearly dependent on thedegree of self-specificity assigned to the pictures by the subjects.While that study did not test for activity changes related to reward,de Grecket al. (9) demonstrated that self-specificity induced ac-tivity changes in these regions (i.e., VTA, VS, VMPFC), thus dem-onstrating an overlap of areas recruited during self and rewardprocessing. Two studies by Enzi et al. (6) and Ersner-Hershfield etal. (7) observed the involvement of other regions, including thepregenual anterior cingulated cortex (PACC), ventrolateral pre-frontal cortex, and the caudate nuclei, during self- and reward-related tasks. Enzi et al. (6), in particular, did this in an interestingway by investigating changes in blood oxygenation level-de-

pendent (BOLD) signaling during closely interleaved trials in-volving reward-related (i.e., a gambling task), self-specific (i.e., apersonal relevance evaluationtask),and control tasks (i.e.,deter-mining the orientation of a visual stimulus). Finally, a recentstudy in a group of young adults (aged 1524) found that the

social aspect of being liked by ones peers activated reward-related and self-specific (i.e., midline) regions. In particular, acti-vations were noted in VS, midbrain (in an area corresponding tothe VTA), VMPFC, posterior cingulate cortex (including retro-splenial cortex), amygdala, and insula/opercular cortex, while

subjects showed greater activation in the VMPFC and amygdalain response to being liked by those whom they regarded highly(21).

Given that nonhuman animal research has largely not ex-plored the concept of self (beyond some work on self-recogni-tionand awareness, e.g., see [2224]), it is challenging to directlycompare reward-related processing and behaviorwhich hasbeen investigated extensivelywith that of the self. Althoughtested indirectly, it may be useful to interpret some animal stud-ies investigating reward-related processing as having self-spe-cific components. To help illustrate this point, consider the fol-lowing analogy: the grand piano has high personal relevance forthe pianist in the same way that a running wheel may have highpersonal relevance for a rodent; this can be determined, for

instance, by assessing the time spent (behavioral) and physio-logical responses during the free pursuit of such things. Thoughone could argue that the pianist can be directly asked about thisrelevance, while the rodent cannot, evidence indicates that self-specificand reward-related associations need not requireaware-

Figure 1. Relationship models of self and reward. The integration model (1) describes a relationship where self-specific and reward-related processing arenearly identical, whereas the segregation model (2) describes a relationship where they are nearly unrelated. The parallel processing model (3) describes arelationship whereby the various aspects of self (e.g., self-specificity) are processed in parallel (i.e., temporally and spatially in an interconnected way) withlower order reward-related processingsuggesting that this relationship combines elements of both the integration and segregation models.

Table 1. Self and Reward in Psychiatric Disorders

Psychiatric Disorder Role of Self and Reward References

Addiction Self-specific stimuli produced BOLD signals in reward circuitry (e.g., VTA, VMPFC, and VS) in healthy subjects;

alcoholics and pathological gamblers displayed normal reward circuitry activity during a reward-related

task but no changes during a self-specific task (i.e., the judgment of stimuli as self- or nonself-specific).

(9,32,33)

Depression Human and animal data of depressive-like behaviors show increased resting state activity in midline regions

(e.g., PACC, VMPFC, VS) that are also implicated in self-specificity and reward-related processing. Reduced

responding of reward-related and self-specific areas (e.g., VS, DMPFC) to positive stimuli in depressedsubjects has also been noted and is in line with the increase in self-focus, or ruminations, seen in this

disorder. It appears as though GABAergic and glutamatergic functioning may play an important role.

(6772)

Schizophrenia Patients show inappropriately strong VS activations to neutral stimuli in an aversive learning paradigm.

Behaviorally, they also show increased temporal discounting and decreased motivation (or wanting),

although their subjective experience of rewards may be intact.

(7376)

Borderline Personality Patients show no PFC (e.g., OFC, DLPFC, anterior cingulate) responses during a reward task and reduced

(e.g., OFC) or increased (e.g., anterior cingulate) responding during a punishment task over healthy

control subjects. Interestingly, a separate study found that these patients show hyperactivity in OFC and

anterior cingulate to both self-specific (or autobiographical) and neutral stimuli, suggesting a deficit of

selective activation.

(77,78)

BOLD, blood oxygenation level-dependent; DLPFC, dorsolateral prefrontal cortex; DMPFC, dorsomedial prefrontal cortex; GABA, gamma-aminobutyricacid; OFC, orbitofrontal cortex; PACC, pregenual anterior cingulated cortex; PFC, prefrontal cortex; VMPFC, ventromedial prefrontal cortex; VS, ventralstriatum; VTA, ventral tegmental area.

2 BIOL PSYCHIATRY 2011;xx:xxx G. Northoff and D.J. Hayes

www.sobp.org/journal


3/7

ness (as noted below), suggesting that behavioral and physio-logical measures are invaluable in this regard.

With that in mind, many animal studies have demonstratedthatVS neurons fire to, and in anticipation of, reward-related stimuli(25). However, there is also a growing body of literature indicatingthat representations of reward that may be personally relevant tothe organism produce activations independent of reward receiptalone (e.g., organism-specific reward-related memories). For exam-

ple, one study in rats found that the reactivation of memory tracesnot only activates feature- and context-specific information butalso contains a value-associated component, indicatingthat activa-tions associated with the memories of personally relevant stimuliare tied directly to reward-relatedprocessing (26). In another study,local field potential oscillations in the rat VS were investigated (27).Results indicated distinct oscillations associated with reward-re-lated memories (7085 Hz) versus those associated with rewardreceipt and reward-related movement initiation (4555 Hz). Theseresults are in line with the other human data, noted above, andsuggest that memories for individual experiences directly involvereward-related processing.

Doesthismeanthatourselfisthusnothingbutreward?Wehaveto be careful with this conclusion because the empirical support is

not uncontroversial. For instance, many studies have also uncov-ered evidence that may be taken to support a nonintegrative mod-elasoutlined in the segregation model section below. Conceptu-ally, the integration model leads to a consideration of the conceptof self in terms of value-awareness. To this, one could logicallyargue that value assignmentdoes not require an explicit awarenessof value, because the former does not necessarily entail awarenessor consciousness and also occurs at an implicit (i.e., unconscious)level (28,29). In contrast, the concept of the self, as describedthroughout the literature, seems to be linked to consciousness(however, as demonstrated below, the self may also be developedand maintained implicitly/without awareness).

Together, the results from human and nonhuman animal stud-ies show some support for the integration model, given the possi-

bility that at least some aspects of self-specific and reward-relatedprocessing occur through integrated mechanisms. Nonetheless,there is also evidence to the contrary. This issue is further amplifiedby considerations of potential differences between the self andreward on the conceptual level. Hence, we turn to the possibilitythat the self and reward are largely separate entities by outliningsome support for the segregation model.

Second Model: Segregation Between Self and RewardContrary to theintegrationmodel, thesegregationmodelof self

and reward claims that our self-specific processing is almost en-tirelyindependent of reward-relatedprocessing. This suggests thatself-specific stimuli are really processed independently from thegeneral valuation processes associated with all self- and nonself-specific stimuli. Value assignment and self-specificity assignmentmaythen be regarded as differentprocesses that areregionally andtemporally segregated.

Evidence from both human imaging and nonhuman animalstudies appears to support key aspects of the segregation model.Forinstance,thestudybydeGrecketal. (9), notedabove, pointsoutsome differences in the duration of the BOLD response associatedwith self-specific and reward-related processing in humans. Basedon raw data analysis (rather than model-based analysis), the au-thors observed sustained BOLD signals during the processing ofself-specific stimuli in the above-mentioned reward circuitry (e.g.,VTA, VMPFC, and VS), while reward-related stimuli induced shorterandmore phasic signals in the same regions. This finding, however,

must be considered preliminary because of the sluggishness of theBOLD response,whichcurrently make it difficult to associateneuraltimings with precision (for an in-depth discussion on related meth-odological issues, the reader is referred to [30,31]).

Nonetheless, this group went on to further substantiate thepossible dissociation between self-specific and reward-relatedpro-cessing in psychiatric patients (Table 1; refer to Supplement 1 for amore in-depth discussion). They observed that while both detoxi-

fied alcoholic patients and pathological gamblers displayednormalneural activity changes in VMPFC, VTA, and VS during a reward-related task, they neverthelessshowed no activity changes in thesesame regions during a self-specifictask (i.e.,the judgment of stimulias self- or nonself-specific) (32,33). The assumption of a regionaldifference between self-specific and reward-related processing isfurther supported by Enzi et al. (6). In this study, the authors foundthat regions such as the insula, premotor cortex, and the su-pragenual anterior cingulate cortex were exclusively associatedwith the processing of self-specific stimuli as distinguished fromreward. Taken together, these findings in humans do indicate apossible dissociation between self and reward either within thesame regions (e.g., reward-related circuitry) or across different re-gions (e.g., the insula or premotor cortex).

There may also be some evidence to support the segregationmodel in studies of nonhuman animals. For instance, it could beargued that the work by Berridge et al. (15) on differentiating be-tween various components of reward, such as liking, wanting, andlearning are also investigating brain-related activity for stimuli thatare both self-specific and reward-related. This group has presentedstrong evidence of a dissociation between wanting (i.e., incentivesalience or the motivation to approach reward-related stimuli,which may involve self-specific processing) and liking (which re-flects the hedonic impact of rewarding stimuli) circuitry. They haveshown a rostrocaudal spatial separation between the circuitry ofwanting and liking in both the VS (34,35) and the ventral pallidum(36). Additionally, the ventral tegmental area-substantianigra com-plex (37) and amygdala (38) have also been implicated in this re-

gard. On the human side, Knutson et al. (39) developed the mone-tary incentive delay task, which includes an anticipation period aspart of a reward task; interestingly, the anticipation period doesindeed induce strong activity in the typical reward regions (e.g.,VTA, VS, VMPFC) even before the reward is actually received (40).Whether such an anticipatory period modulates or even simulatesactivity related to what is described by Berridge et al. (15) as want-ing in animals, however, remains open.

Congruent with the data from alcoholic patients and patholog-ical gamblers above, some stimuli (i.e., drug-associated cues) takeon high incentive salience (wanting) roles but have low reward(liking) values in animals in addicted-like states (what Robinson andBerridge [41] have termed incentive sensitization). These behav-ioral observations (e.g., via drug self-administration) are supportedby theevidenceimplicating,for instance,dysregulation of theVS inmediating the abnormal wanting, but not the liking, component inanimals trained to associate cues with the reception of drugs ofabuse (42,43). This incentive sensitization results in what theseauthors refer to as an irrational wanting for something that is notcognitively wanted (15). Importantly, this dissociation is in line withthat seen in alcoholic patients and pathological gamblers showingnormal VS function in the reward-related task but abnormal re-sponding during the self-specific task (9,33).

As with the integration model, there are some conceptual con-siderations and caveats regarding the segregation model. For in-stance, it is tempting to speculatethat there maybe no self-specificcomponent to liking(i.e., the hedonic impact of rewards), giventhat

G. Northoff and D.J. Hayes BIOL PSYCHIATRY 2011;xx:xxx 3



4/7

it does notrequire consciousawareness (28,44). Onthe other hand,there may be a self-specific component to wanting, given that thisconcept reflects the level of importance placed on external stimuliby the organism. Additionally, contrasting the results from healthyversus addicted-like animals and humans suggests that at somelevel self-specific and reward-related processing are segregatedand can become dysregulated (as reflected in addicted-like states).Nonetheless, while there is strong support for the distinction be-

tweenliking and wanting regarding reward-related processingandthe data from these studies may be in line with the presence ofself-specific processing, there is currently no direct evidence for (oragainst) the involvement of a self-specific component. Hence, weconsider a third model, which attempts to reconcile thesupport foreach the integration and segregation models.

Third Model: Parallel Processing of the Self Along the RewardContinuum

The parallel processing model aims to account for all the obser-vations noted above by considering that different aspects of self-specific processing may occur in parallel with aspects of reward-related processing. As alluded to above, the full spectrum ofreward-related processing involves different neural substrates (in-cluding both cortical and subcortical areas), as well as a range oftemporal responses (early vs. later phase responses). Intuitively, itcouldbe hypothesized thatvariousaspectsof the selfare processedin parallel along a continuum of reward-related processing (Figure1). Next, we will weigh theevidenceand conceptual considerationsfor this model by focusing on (what we have termed here) thelower-order end of the continuum. At this extreme, self-specificprocessingcould be considered the ground uponwhich the assign-ment of value to external stimuli becomes possible. This lowerorder designation may reflect the basic processing that dominatesmainly (but perhaps not exclusively) in subcortical regions andearlier in time. Although self-specific processing may take place atother levels of the continuum, we have not considered this here indepth (although the section ends with a brief note on the higherorder end).

From an anatomical viewpoint, imaging studies demonstratethe involvementof subcortical regions likethe periaqueductalgray,the VS, putamen/caudate, the tectum, and the VTA in self-specific-ity assignment (6,8,9,4547). This is particularly seen in studies thatrely on passive perception, as opposed to more cognitive involve-mentin whichsubjects haveto explicitly judge the stimulusdegreeof self-specificity. Conversely, reward and value assignment do notonly implicate subcortical regions like the VS and the VTA but alsovarious cortical regions including the VMPFC, the lateral prefrontalcortex, and the parietal cortex (48). Hence, the imaging findings donot support the distinction between reward- and self-specific pro-cessing strictly along subcortical and cortical borders.

Indeed, the observed overlap between reward- and self-specificprocessing in the core regions of the reward circuitry (e.g., VS, VTA,VMPFC, and PACC) in humans (6,7,32,33,49) and in animals(26,34,37) further underscores that the concepts of reward and selfcannot be strictly delineated through gross neuroanatomy. As an-otherexample of this,self-specificity assignment appearsto involvespecific regions like the insula that have been implicated in theprocessing of interoceptive stimuli from ones body (6,5054). Thismay distinguish it from reward-related processing, which may bemore focused on exteroceptive rather than interoceptive stimuli.However, there is evidence of its involvement in reward-relatedprocessing, supported by the recruitment of the insula in sociallyrelated reward tasks (55,56), aswellas by its putative role in subjec-tive drug craving (57). However, the exact relationship between

value and self-specificity assignment in the insula (and other re-

gions), as well as the impact of interoceptive processing, remains

unclear(seeSupplement1formorediscussionontheroleofintero-

ception).More support comes from theconsideration that thesamebrain

regions may be implicated in different processes at different times.

Schultz (14,58) has argued that the activities of dopamine neurons

may result in different effects at different time scales. For instance,

short phasic increases (reward prediction error), more sustainedphasic responses (reward uncertainty), and slower depressions

(aversive signaling) are each associated with different aspects of

processing. Additionally, slow tonic dopaminergic changes across

time intervals of around 20 to 30 minutes appear to broadly affectmovements, cognition, and motivation. In a similar fashion, neural

activity within regions may mediate different functions (e.g., re-

lated to rewardand self) at differenttime scales. In-depth investiga-

tion of these time scales and the other transmitters likely involved

(e.g., gamma-aminobutyric acid, glutamate, and serotonin) may bebest undertaken in nonhuman animals. For this, however, we must

first better operationalize aspects of reward- and self-specific pro-

cessing in animals (see Supplement 1 for more discussion). Al-

though these are methodologically difficult to discern in human

imaging paradigms, this would help to explain why there is neuraldissociation between selfandreward,even within the sameregions

(9,32,33). If this holds, one could argue that what we call value

assignment operating within the millisecond-to-second range may

reflect a temporal dissociation from our concept of self-specificity.This, however, remains to be demonstrated (see Supplement 1 for

more discussion on the role of temporal coding with regard to the

self).

In additionto timing, thebroaderquestionof neuralcoding may

also need to be discussed. Reward has also been associated withpredictive coding, the notion that the to-be-expected stimulus is

anticipated and then matched and compared with the actual in-

coming stimulus (59). The difference between anticipated and ac-

tual stimulus may then reflect what is called the prediction error,which is assumed to determinethe degreeof neuralactivityrelatedto that particular stimulus. This has been quite convincingly dem-

onstratedin thecase of reward(60,61), while theissue of predictive

coding has not been addressed in the context of the self. If reward-

and self-specific processing do indeed overlap in terms of regions

and neural activity, one would expect the neural activity related tothe self to also be determined by the prediction error signaling the

degree of mismatch in self-specificity between the anticipated and

actual stimulus. This, however, would mean that neural activity in,

for instance, cortical midline structures during the perception ofself-specific stimuli may rather reflect a difference-based signal,

e.g., a difference in anticipated and actual self-specificity, rather

than self-specificity per se.

What do these empirical findings entail for the concept of theself as characterized by self-specific processing? The concept ofself-specific processing sketched here must be distinguished from

the one of self-referential processing that describes the organisms

reference to the results of its own stimulus processing (2). The

concept of self-referentialprocessing shiftsthe focusfrom the more

basicand lowerorder end internal-external organism-environmentrelationship to a purely internal and therefore higher order end

organism-organism relationship (Figures 1 and 2). Hence, the con-

cept of self-referentialprocessing pertains functionally to the meta-representation of what is constituted on the level of self-specific

processing. Self-referential processing may then be considered the

higher order end, whereby self-specific processing occurs at the




5/7

lower order end with both being potentially mediated by value

assignment.Finally, one should also be aware that the concept of reward iscurrently very different from what it was just a decade ago. While,formanyyears, theconceptof rewardwasoftenintimately linkedtopleasure (62), recent experiments have described a heterogeneousarray of related functions involving pleasure/hedonia/liking, want-ing/incentive salience, seeking, instrumental learning, condition-ing, prediction, and behavioral activation (12,58,63,64). While re-ward as pleasure is too general in some way, it may also be toospecific by restricting reward to motivational features of positivereinforcers (11). Another concept often closely associated with re-ward is salience, which refers to the ability of stimuli to inducereallocation of cognitive resources beyond its mere value (see, forinstance [65,66]). Defined in terms of resource allocation, saliency

may provide a conceptual bridge from reward to the self: stimulirelated to the self are highly relevant and important and thereforemay recruit and allocate more sensory, affective, and cognitiveresources compared with nonself-related ones. However, this re-mains purely hypothetical at this point and awaits empiric testing.Finally, reward may even be further complicated by consideringaspects suchas the linkage of interoceptive andexteroceptive stim-uli andtemporal extension, as may be necessary when consideringrewarddirectly with theself.Overall, this discussion suggests that itmay not only be the field of self that benefits from turning towardthe field of reward but also vice versa.

Conclusions

The discussion of different possible models revealed that selfand reward can neither be considered identical (i.e., integrationmodel) nor clearly separate identities (i.e., segregation model). In-stead, their relationship seems to be rather complex with multipleinteractions across a continuum that may be evident neurally, psy-chologically, and conceptually. Unfortunately, these results remaintentative, given that the focus on comparisons of self- and nonself-specific stimuli has generatedcontroversy and unclear conclusions.Nonetheless, this makes the future investigation of the interactionbetween self and reward an exciting field where both may benefitfrom one another.

By intertwining the fields of self and reward and orienting to-ward what the empirical data tell us, a far clearer understanding ofeach concept mayemerge.As is already startingto occur in thefield

of reward, this mayultimatelyleadto thedeterminationof theexactprocesses and psychological functions associated with the con-cepts of self and reward (for example, involving interoceptive-ex-teroceptive linkage). This increased interaction and reciprocationbetween the fields across various levels will undoubtedly lead toexciting new hypotheses and discoveries that will foster the devel-opment and refinement of these concepts.

GN holds a Canada Research Chair for Mind, Brain Imaging andNeuroethics, as wellas an EdithJacobsonLow-BeerFoundation-Cana-dian Institutes of Health Research Michael Smith Chair in Neurosci-ences and Mental Health. DJH holds a Postdoctoral Fellowship fromthe Canadian Institutes of Health Research.

The authors report no biomedical financial interests or potentialconflicts of interest.

Supplementary material cited in this article is availableonline.

1. Northoff G, Bermpohl F (2004): Cortical midline structures and the self.TrendsCogn Sci8:102107.

2. Northoff G, Heinzel A, de Greck M, Bermpohl F, Dobrowolny H, Pank-seppJ (2006): Self-referential processing in our braina meta-analysisofimaging studies on the self. Neuroimage 31:440457.

3. NorthoffG, PankseppJ (2008): Thetrans-speciesconceptof self andthesubcortical-cortical midline system. Trends Cogn Sci12:259264.

4. Gillihan SJ, Farah MJ (2005): Is self special? A critical review of evidencefromexperimentalpsychology and cognitive neuroscience. Psychol Bull131:7697.

5. Legrand D, Ruby P (2009): What is self-specific? Theoretical investiga-tion and critical review of neuroimaging results. Psychol Rev116:252282.

6. Enzi B, de Greck M, Prosch U, Tempelmann C, Northoff G (2009): Is ourself nothing but reward? Neuronal overlap and distinction betweenreward and personal relevance and its relation to human personality.PLoSONE4:e8429.

7. Ersner-Hershfield H, Wimmer GE,KnutsonB (2009): Saving for the futureself: Neural measures of future self-continuity predict temporal dis-counting. Soc Cogn Affect Neurosci4:8592.

8. PhanKL, TaylorSF, Welsh RC,Ho SH,BrittonJC, Liberzon I (2004): Neural

correlatesof individual ratings of emotional salience: A trial-relatedfMRIstudy. Neuroimage 21:768780.

9. de Greck M, Rotte M, Paus R, Moritz D, Thiemann R, Proesch U, et al.(2008): Is our self based on reward? Self-relatedness recruits neuralactivity in the reward system. Neuroimage 39:20662075.

10. Metzinger T, Gallese V (2003): The emergence of a shared action ontol-ogy: Building blocks for a theory. Conscious Cogn 12:549571.

11. Salamone JD (2006): Will the last person who uses the term rewardplease turnout the lights? Commentson processesrelatedto reinforce-ment, learning, motivation and effort. Addict Biol11:4344.

12. Salamone JD (2007): Functions of mesolimbic dopamine: Changingconcepts and shifting paradigms. Psychopharmacology (Berl) 191:389.

13. Montague PR (2007): Neuroeconomics: A view from neuroscience.FunctNeurol22:219234.

14. Schultz W (2006): Behavioral theories and the neurophysiology of re-ward.Annu Rev Psychol57:87115.

15. Berridge KC, Robinson TE, Aldridge JW (2009): Dissecting componentsof reward: liking, wanting, and learning.CurrOpin Pharmacol9:6573.

16. StephensDN, DukaT, Crombag HS,CunninghamCL, Heilig M, Crabbe JC(2010): Reward sensitivity: Issues of measurement, and achieving con-silience between human and animal phenotypes. Addict Biol 15:145168.

17. Robbins TW, Everitt BJ (2007): A role for mesencephalic dopamine inactivation: Commentary on Berridge (2006). Psychopharmacology (Berl)191:433437.

18. Montague PR, Berns GS (2002): Neural economics and the biologicalsubstrates of valuation. Neuron 36:265284.

19. MontaguePR, King-Casas B, CohenJD (2006): Imaging valuationmodelsin human choice.Annu Rev Neurosci29:417448.

20. Panksepp J, Northoff G (2009): The trans-species core SELF: Theemergence of active cultural and neuro-ecological agents through

Figure 2. Different functionalprocesses and their hypothesized neural tim-ing. Hypothesized continuous relationship between reward- and self-re-lated processing over time. Note that theexamplesand related time courseare hypothesized and are for illustrative purposes only.




6/7

self-related processing within subcortical-cortical midline networks.Conscious Cogn 18:193215.

21. Davey CG, Allen NB, Harrison BJ, Dwyer DB, Yucel M (2010): Being likedactivates primary reward and midline self-related brain regions. HumBrain Mapp 31:660668.

22. Gallop GG Jr (1970): Chimpanzees: Self-recognition. Science 167:8687.23. Iriki A, TanakaM, Obayashi S, Iwamura Y (2001): Self-images inthe video

monitor coded by monkey intraparietal neurons. Neurosci Res 40:163173.

24. Plotnik JM, de Waal FB, Reiss D (2006): Self-recognition in an Asianelephant. ProcNatlAcad Sci U S A 103:1705317057.25. Schultz W, Tremblay L, Hollerman JR (2000): Reward processing in pri-

mate orbitofrontal cortex and basal ganglia. Cereb Cortex10:272284.26. Lansink CS, Goltstein PM, Lankelma JV, Joosten RN, McNaughton BL,

Pennartz CM (2008): Preferential reactivation of motivationally relevantinformation in the ventral striatum. J Neurosci28:63726382.

27. vanderMeer MA,RedishAD (2009): Lowandhighgammaoscillations inrat ventral striatum have distinctrelationships to behavior, reward, andspiking activity on a learned spatial decision taskFront Integr Neurosci3:9.

28. Pessiglione M, Schmidt L, Draganski B, Kalisch R, Lau H, Dolan RJ, FrithCD (2007): How the brain translates money into force: A neuroimagingstudy of subliminal motivation. Science 316:904906.

29. Rangel A (2008): Consciousness meets neuroeconomics: What is thevalue of stimulus awareness in decision making? Neuron 59:525527.

30. Bandettini PA (2009): Whats new in neuroimaging methods? Ann N YAcad Sci1156:260293.31. Ekstrom A (2010): Howandwhenthe fMRI BOLD signal relates to under-

lying neural activity: The danger in dissociation. Brain Res Rev 62:233244.

32. de Greck M, Enzi B, Prosch U, Gantman A, Tempelmann C, Northoff G(2010): Decreased neuronal activity in reward circuitry of pathologicalgamblers during processing of personal relevant stimuli. Hum BrainMapp 31:18021812.

33. de Greck M, Supady A, Thiemann R, Tempelmann C, Bogerts B, Forsch-ner L, et al. (2009): Decreased neural activity in reward circuitry duringpersonal reference in abstinent alcoholics--a fMRI study. Hum BrainMapp 30:16911704.

34. Pecina S, Berridge KC (2005): Hedonic hot spot in nucleus accumbensshell: Where do mu-opioids cause increased hedonic impact of sweet-ness?J Neurosci25:1177711786.

35. Smith KS, Berridge KC (2007): Opioid limbic circuit for reward: Interac-tion between hedonic hotspotsof nucleus accumbensand ventral pal-lidum.J Neurosci27:15941605.

36. Smith KS, Berridge KC (2005): Theventralpallidum andhedonic reward:Neurochemical maps of sucrose liking and food intake. J Neurosci25:86378649.

37. Martinez-Hernandez J, Lanuza E, Martinez-Garcia F (2006): Selectivedopaminergic lesions of the ventral tegmental area impair preferencefor sucrose but not for male sexual pheromones in female mice. Eur

J Neurosci24:885893.38. Mahler SV, Berridge KC (2009): Which cue to want? Central amygdala

opioid activation enhances and focuses incentive salience on a prepo-tent reward cue.J Neurosci29:65006513.

39. KnutsonB, WestdorpA, Kaiser E, Hommer D (2000):FMRI visualizationofbrain activity during a monetary incentive delay task. Neuroimage 12:2027.

40. KnutsonB, TaylorJ, Kaufman M, PetersonR, GloverG (2005): Distributedneural representation of expected value. J Neurosci25:48064812.

41. Robinson TE, Berridge KC (1993): The neural basis of drug craving: Anincentive-sensitization theory of addiction. Brain Res Brain Res Rev 18:247291.

42. Wyvell CL, Berridge KC (2000): Intra-accumbens amphetamine in-creases the conditioned incentive salience of sucrose reward: Enhance-ment of reward wanting without enhanced liking or response rein-forcement.J Neurosci20:81228130.

43. Wyvell CL, Berridge KC (2001): Incentive sensitization by previous am-phetamine exposure: Increased cue-triggered wanting for sucrosereward.J Neurosci21:78317840.

44. Seitz AR, Kim D, Watanabe T (2009): Rewards evoke learning of uncon-sciously processed visual stimuli in adult humans. Neuron 61:700707.

45. Schneider F, Bermpohl F, Heinzel A, Rotte M, Walter M, Tempelmann C,et al. (2008): The resting brain and our self: Self-relatedness modulates

resting state neural activity in cortical midline structures. Neuroscience157:120131.

46. QinP, DiH,LiuY,YuS, GongQ,DuncanN, etal. (2010): Anterior cingulateactivity and the self in disorders of consciousness. Hum Brain Mapp31:19932002.

47. Northoff G, Schneider F, Rotte M, Matthiae C, Tempelmann C, WiebkingC, et al. (2009): Differential parametric modulation of self-relatednessand emotions in different brain regions. Hum Brain Mapp 30:369382.

48. Haber SN, Knutson B (2010): The reward circuit: Linking primate anat-

omy and human imaging. Neuropsychopharmacology35:426.49. Ersner-Hershfield H, Garton MT, Ballard K, Samanez-Larkin GR, KnutsonB (2009): Dont stop thinking about tomorrow: Individual differences infuture self-continuity account for saving.Judgm Decis Mak4:280286.

50. CraigAD (2003): Interoception: The sense of thephysiological conditionof thebody. Curr Opin Neurobiol13:500505.

51. Craig AD (2004): Human feelings: Why are some more aware than oth-ers? TrendsCogn Sci8:239241.

52. WiebkingC, Bauer A, de Greck M, DuncanNW, Tempelmann C, NorthoffG (2010): Abnormal body perception andneural activity in theinsula indepression: An fMRI study of the depressed material me. World J BiolPsychiatry11:538549.

53. Craig AD (2009): Emotional moments across time: A possible neuralbasis for timeperception in the anterior insula. PhilosTrans R SocLondBBiol Sci364:19331942.

54. Craig AD (2010): The sentient self. Brain Struct Funct214:563577.

55. Montague PR, King-Casas B (2007): Efficient statistics, common curren-ciesand theproblemof reward-harvesting.Trends CognSci11:514519.56. Sanfey AG, Rilling JK, Aronson JA, Nystrom LE, Cohen JD (2003): The

neural basis of economic decision-making in the Ultimatum Game.Science 300:17551758.

57. Naqvi NH, Bechara A (2009): The hidden island of addiction: The insula.Trends Neurosci32:5667.

58. Schultz W (2007): Behavioral dopamine signals. Trends Neurosci 30:203210.

59. Friston K, KiebelS (2009): Predictive codingunderthe free-energy prin-ciple. PhilosTransR Soc LondB Biol Sci364:12111221.

60. Tobler PN, ODoherty JP, Dolan RJ, Schultz W (2006): Human neurallearningdependson rewardpredictionerrors in theblocking paradigm.

J Neurophysiol95:301310.61. Schultz W (1998): Predictive rewardsignal of dopamine neurons.J Neu-

rophysiol80:127.62. Kringelbach ML, Berridge KC (2009): Towards a functional neuroanat-

omy of pleasure andhappiness. Trends Cogn Sci13:479487.63. Berridge KC, Kringelbach ML (2008): Affective neuroscience of pleasure:

Reward in humans and animals.Psychopharmacology(Berl) 199:457480.64. Ikemoto S, PankseppJ (1999):The roleof nucleusaccumbens dopamine

in motivated behavior: A unifying interpretation with special referenceto reward-seeking. Brain ResBrain ResRev31:641.

65. Zink CF, Pagnoni G, Chappelow J, Martin-Skurski M, Berns GS (2006):Human striatal activation reflects degree of stimulus saliency. Neuroim-age 29:977983.

66. Zink CF, Pagnoni G, Martin-Skurski ME, Chappelow JC, Berns GS (2004):Human striatalresponses to monetary reward depend on saliency.Neu-ron 42:509517.

67. EpsteinJ, PanH, KocsisJH,YangY, ButlerT, ChusidJ, etal. (2006): Lackofventral striatal response to positive stimuli in depressed versus normalsubjects.Am J Psychiatry163:17841790.

68. Northoff G (2007): Psychopathology and pathophysiology of the self indepressionneuropsychiatric hypothesis.J Affect Disord104:114.

69. GrimmS, ErnstJ, Boesiger P, Schuepbach D,HellD, BoekerH, NorthoffG(2009): Increased self-focus in major depressive disorder is related toneural abnormalities in subcortical-cortical midline structures. HumBrain Mapp 30:26172627.

70. WalterM, Henning A, Grimm S, Schulte RF,Beck J, Dydak U, etal. (2009):The relationship between aberrant neuronal activation in the pre-genual anterior cingulate, altered glutamatergic metabolism, and an-hedonia in major depression.Arch Gen Psychiatry66:478486.

71. BermpohlF, WalterM, SajonzB, LuckeC, HageleC, Sterzer P, etal. (2009):Attentional modulation of emotional stimulus processing in patientswith major depressionalterations in prefrontal cortical regions. Neu-rosci Lett463:108113.

72. Alcaro A, Panksepp J, Witczak J, Hayes DJ, Northoff G (2010): Is subcor-tical-cortical midline activity in depression mediated by glutamate and




7/7

GABA? A cross-species translational approach. Neurosci Biobehav Rev34:592605.

73. JensenJ,WilleitM, ZipurskyRB, SavinaI,SmithAJ,Menon M, etal. (2008):The formation of abnormal associations in schizophrenia: Neural andbehavioral evidence. Neuropsychopharmacology33:473479.

74. Heerey EA, Gold JM (2007): Patients with schizophrenia demonstratedissociation between affective experience and motivated behavior. J

Abnorm Psychol116:268278.75. HeereyEA, Robinson BM,McMahonRP, Gold JM (2007): Delay discount-

ing in schizophrenia. Cogn Neuropsychiatry12:213221.

76. Juckel G, Sass L, Heinz A (2003): Anhedonia, self-experience in schizo-phrenia, and implications for treatment. Pharmacopsychiatry36(suppl3):S176S180.

77. Vollm B, Richardson P, McKie S, Elliott R, Dolan M, Deakin B (2007):Neuronal correlates of reward and loss in Cluster B personality disor-ders: A functional magnetic resonance imaging study. Psychiatry Res156:151167.

78. Schnell K, Dietrich T, Schnitker R, Daumann J, Herpertz SC (2007): Pro-cessing of autobiographical memory retrieval cues in borderline per-sonality disorder.J Affect Disord97:253259.



Is Our Self Nothing but Reward.pdf

Documents

Transcript of Is Our Self Nothing but Reward.pdf