1198 VOLUME 18 | NUMBER 9 | SEPTEMBER 2015 NATURE NEUROSCIENCE NEWS AND VIEWS The interpretation of the study by Guo et al. 4 also opens an interesting debate on the complex relationship between structural spine changes and functional plasticity, in the form of LTP. In fact, the authors hypothesize that spines repre- sent a substrate for structural plasticity that does not necessarily correlate with functional plas- ticity. Indeed, the loss of LTP associated with lack of endogenous dopamine was not directly related to the observed structural spine changes. Conversely, the authors attribute to LTP the ability to stabilize newly formed spines. The analysis of dynamic changes of spine morphology could have implications not only for the pathophysiology of PD, but also for the motor and behavioral side effects related to dopamine replacement therapy of this neu- rodegenerative disorder. After a few years of treatment, l-DOPA causes a series of hyper- kinetic motor symptoms called l-DOPA– induced dyskinesias that limit the use of this therapy. These dyskinesias are thought to be caused by loss of LTP reversal at cortico-striatal synapses 7 . This latter event, also referred to as depotentiation, is usually induced by low- frequency stimulation (1–2 Hz). It represents the ability of an already potentiated synapse to return to control levels and allows the erasure of unessential memory information. Nevertheless, alterations in spine dynam- ics at the level of motor cortex could just as well contribute to this disabling phenom- enon. Similarly, aberrant plasticity at cortical synapses may be responsible for the impulse control disorders observed during dopamin- ergic treatment. Impulse control disorders, including compulsive gambling, buying, sexual behavior and eating, are increasingly recognized as serious psychiatric complica- tions in PD 8 , and they may be the result of an anomalous interaction between dopamine and glutamate at the level of the cortical spines. From the clinical point of view, three main aspects should be considered in the interpreta- tion of most of the data resulting from PD ani- mal models, including the study by Guo et al. 4 . First, PD is a slow neurodegenerative disorder in which dopaminergic denervation occurs over decades. By contrast, dopaminergic denervation is induced in most preclinical studies by acute neurotoxic lesions. Structural plasticity may dif- fer in these two situations. Second, aging is the greatest risk factor for the development of PD, as well as other neurodegenerative disorders 9 . Yet most experimental studies, including that of Guo et al. 4 , use young animals, in which plastic events may not be influenced by the different molecular environment that characterizes the aging brain Third, PD is more than just dopamine loss 10 . The widespread, multisystem nature of the neu- rodegeneration that characterizes PD leads to the involvement of different neurotransmitters, including acetylcholine, serotonin and nora- drenaline 11 , whose modulation could act in concert with dopamine loss to influence both structural and functional plasticity. Nevertheless, imaging of spine turnover coupled to electrophysiological analyses of synaptic plasticity and behavioral investigation represents a powerful approach to clarifying the role of the cortex in neurotransmitter-related disorders. Future studies implementing a simi- lar approach in the deeper basal ganglia nuclei might allow the precise description of an inte- grative model of motor control and subcellular mechanisms underlying movement disorders. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. 1. Obeso, J.A., Rodriguez-Oroz, M.C., Stamelou, M. & Bhatia, K. Lancet 384, 523–531 (2014). 2. Calabresi, P., Picconi, B., Tozzi, A., Ghiglieri, V. & Di Filippo, M. Nat. Neurosci. 17, 1022–1030 (2014). 3. Day, M. et al. Nat. Neurosci. 9, 251–259 (2006). 4. Guo, L. et al. Nat. Neurosci. 18, 1299–1309 (2015). 5. Yagishita, S. et al. Science 345, 1616–1620 (2014). 6. Ueno, T. et al. Neurobiol. Dis. 64, 142–149 (2014). 7. Calabresi, P., Ghiglieri, V., Mazzocchetti, P., Corbelli, I., Picconi, B. Levodopa-induced plasticity: a double- edged sword in Parkinson’s disease? Philos. Trans. R. Soc. Lond. B Biol. Sci. 370, 20140184 (2015). 8. Weintraub, D., David, A.S., Evans, A.H., Grant, J.E. & Stacy, M. Mov. Disord. 30, 121–127 (2015). 9. Collier, T.J., Kanaan, N.M. & Kordower, J.H. Nat. Rev. Neurosci. 12, 359–366 (2011). 10. Lang, A.E. & Obeso, J.A. Lancet Neurol. 3, 309–316 (2004). 11. Hall, H. et al. Brain 137, 2493–2508 (2014). Character studies Ming Hsu & Adrianna C Jenkins How do individuals attribute dispositional properties, or traits, to others? A study suggests that associative learning processes underlie aspects of trait learning at both neural and behavioral levels. Ming Hsu and Adrianna C. Jenkins are in the Haas School of Business and Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, California, USA. e-mail: mhsu@haas.berkeley.edu In O. Henry’s “The Gift of the Magi ”, a young wife sells her prized hair to buy her husband a chain for his gold watch, while the husband sells the watch to buy her expensive combs. Thus, unbeknownst to the other, each is left with a gift that neither can use. One possible takeaway is that the gifts failed miserably and that the couple should have consulted each other before their purchases. But most of us focus instead on the husband and wife’s generous dispositions. Beginning with Heider 1 , how perceivers attri- bute dispositional properties, or traits, to others has been among the most enduring questions in psychology 2 . In recent years, there has been increasing interest in approaching this ques- tion using cognitive neuroscience techniques. However, despite important advances 3 , we remain far from a mechanistic understanding of how particular brain regions enable trait infer- ence, particularly regarding the computations essential to binding high-level theories of social cognition to the underlying neurobiology. In a functional magnetic resonance imaging (fMRI) study reported in this issue of Nature Neuroscience, Hackel, Doll and Amodio 4 take an important step in this direction. Using an innovative combination of ideas and tools from social psychology, economics and cog- nitive neuroscience, they offer neural evi- dence that associative learning processes are involved in making inferences about traits. Specifically, the authors conducted a study in which participants interacted repeatedly with eight different partners: four purported human participants and four slot machines. On each trial, participants chose to interact with one of two human (or slot machine) counterparts. The chosen counterpart, who had been endowed with a certain number of points on that trial, then shared some proportion of those points with the participant. Critically, targets varied orthogonally in terms of the average magnitude of their starting endow- ment (reward) and the average proportion of the endowment that was shared with the par- ticipant (generosity), enabling the authors to dissociate signals associated with trait learning from those associated with reward processing. Consistent with the idea that trait learning engages associative learning processes, BOLD (blood oxygen level-dependent) responses of the npg © 2015 Nature America, Inc. All rights reserved.