The head and tail of caudate nucleus code flexible and stable reward value of visual objects

Being able to estimate the reward value of visual objects is a crucial factor in guiding ones behavioral choices. What makes this task even more challenging is that the reward value is often not fixed but can vary quickly, making it necessary for there to be flexibility to take into account the immediate reward history in addition to the stable reward value that has been learned over the longer term. The underlying neural mechanisms have remained a topic of speculation. Existence of two parallel reward-value processing mechanisms, one processing flexibly short-term reward value and another holding the longer-term stable reward value of objects has been hypothesized, however, empirical support for this hypothesis has been lacking.

In their recent study, Drs. Kim and Hikosaka (2013) used combined single-neuron recording and temporal inactivation methods in non-human primates to investigate the roles of distinct caudate nucleus areas in determination of flexible and stable reward values of visual stimuli. They observed that, during behavioral tasks wherein monkeys looked more at objects with high than low value, neuronal firing recorded from head of the caudate nucleus coded reward value flexibly and neurons in the tail of the caudate nucleus coded for the longer-term stable reward value. Temporary inactivation of these two caudate nucleus subregions corroborated the findings obtained in the single-neuron recordings.

These very important and exciting results suggest that there indeed are two parallel neural systems coding for reward value of objects, one enables flexible coding of reward value when there is short-term volatility in value, and another mechanism enables holding and appreciating the stable reward values of objects. In addition to shedding light on the neural basis of reward-value processing under different types of task conditions, these findings offer hypotheses and insights into the neural basis of specific deficits that have been documented in various basal ganglia disorders, as also briefly discussed by the authors in their paper.

Reference: Kim HF, Hikosaka O. Distinct basal ganglia circuits controlling behaviors guided by flexible and stable values.  Neuron (2013) e-publication ahead of print. http://dx.doi.org/10.1016/j.neuron.2013.06.044


Unexpectedness of both observed errors and successes activates the dorsomedial prefrontal and rostral cingulate cortex in humans

It has been sometimes said that the ability to predict what is going to happen next is the primary task that the human brain needs to accomplish (e.g., perhaps the reason that ability to form memories of past events ever developed was solely due to the need to be able to predict the future). Indeed, when observing others, there typically are few surprises, and unexpected acts robustly catch one’s attention to figure out what is taking place. Activation of dorsomedial prefrontal cortical areas together with rostral cingulate cortex has been associated in previous studies with both detection of errors (e.g., when observing someone fail on a task) and observation of surprising events. Whether the responses seen in these areas are more due to unexpectedness or erroneousness of observed actions has, however, remained as an unresolved issue.

In their recent study, Dr. Anne-Marike Schiffer et al. (2013) studied whether responses in the aforementioned brain areas are more due to unexpectedness or erroneousness of observed actions. They presented video clips, shot from a first-person perspective, of an actress making sailing, fishing, and climbing knots. The videos were edited so that both unexpected failures and unexpected successes were observable, as also validated in a separate behavioral experiment carried out in the volunteers who all were sufficiently skilled in making the knots themselves. The movie clips were then shown to the volunteers during functional magnetic resonance imaging. The results indicated an area encompassing medial prefrontal cortex and rostral cingulate cortex that responded to both correct and erroneous knot-tying actions that were unexpected.

These very important and interesting results suggest that, at least to some extent, previously observed error-related responses in dorsomedial prefrontal cortex and rostral cingulate gyrus could have been due to unexpectedness of the errors. Based on their findings, the authors further bring up the interesting possibility that an unexpectedness signal in the dorsal rostral cingulate gyrus could serve the purpose of adjusting internal models that help predict flow of actions. Overall, this study is a very nice demonstration of how behavioral and neuroimaging experiments can be combined to advance our understanding of the neural basis of cognitive functions.

Reference: Schiffer A-M, Krause KH, Schubotz RI. Surprisingly correct: unexpectedness of observed actions activates the medial prefrontal cortex. Human Brain Mapping (2013) online e-publication ahead of print. http://dx.doi.org/10.1002/hbm.22277


Brain regions processing complex acoustic features across different musical genres

Music is a fundamental and highly interesting aspect of humanity. The neural basis of music perception has been studied for the most part with relatively simplified stimuli isolating a given element of music, such as by presenting short sound sequences that form tonality or rhythm, and observing which brain areas exhibit responses to such stimulation. Over the last few years, there has been an emerging trend, enabled by developments in non-invasive neuroimaging technology and data analysis methods, towards utilization of naturalistic stimuli during neuroimaging, including free listening of music. What has been wanting, however, are studies looking at which brain areas are consistently activated by musical features across different musical pieces and genres during free listening conditions.

In their recent study, Alluri et al. (2013) presented healthy volunteers musical pieces of various genres that included both instrumental music and music with lyrics during functional magnetic resonance imaging. Musical features were then extracted by automated algorithms included in the so-called MIR toolbox that the authors have developed previously. These complex acoustic feature time series were then used as regression models to predict voxel-wise brain hemodynamic activity recorded during music listening. Cross-validation was used across musical genres and two different subject populations to map areas that respond consistently to the musical complex acoustic features.

It was shown that brain activity can be predicted by the musical complex acoustic features in the auditory, limbic, and motor regions of the brain, as well as in orbitofrontal regions that have been previously associated with evaluative appraisal and not during free music listening per se. Cross-validation identified a region in right superior temporal gyrus that included planum polare and Heschl’s gyrus as the core structure that processes complex acoustic features across musical genres. These highly exciting findings will help pave way for further neuroimaging studies into the neural basis of music processing under naturalistic free music listening conditions.

Reference: Alluri V, Toiviainen P, Lund TE, Wallentin M, Vuust P, Nandi AK, Ristaniemi T, Brattico E. From Vivaldi to Beatles and back: predicting lateralized brain responses to music. Neuroimage 83 (2013) 627-636. http://dx.doi.org/10.1016/j.neuroimage.2013.06.064


Psychophysics of spatial hearing and the underlying neural mechanisms in humans nicely reviewed

Localization of sound sources is a complicated challenge for the human brain since the auditory system, unlike the visual one, lacks direct correspondence between sound source locations and sensory receptive fields. In their recent review article, Dr. Jyrki Ahveninen et al. (2013) provide a comprehensive review of what is known about the psychophysics of sound localization and the current understanding of the underlying cortical mechanisms as elucidated by neuroimaging studies.

Both animal models and more recently non-invasive neuroimaging studies in humans have suggested a special role in auditory spatial processing for cortical areas that reside posterior to the primary auditory cortex, including planum temporale and posterior superior temporal gyrus, however, both the precise underlying neural mechanisms have remained in many ways an unresolved puzzle in cognitive neuroscience. The most significant outstanding questions are laid out in the paper, which is a good read for anyone interested in the cognitive neuroscience of spatial hearing.

Reference: Ahveninen J, Kopco N, Jaaskelainen IP. Psychophysics and neuronal bases of sound localization in humans (2013) Hearing Research, e-publication ahead of print. http://dx.doi.org/10.1016/j.heares.2013.07.008