One year of aerobic exercise enhances white matter integrity in prefrontal and temporal areas in older adults

Aging-related decline in cognitive function is a rapidly increasing societal challenge in nations where the population is aging rapidly such as in European Union countries. With normal aging there are a number of changes in brain structure and function, and degeneration of cerebral white matter (that is composed of connections linking brain areas together) has been consistently observed to take place with aging. This degeneration of brain connectivity has been further observed to correlate with aging-related decline in cognitive functioning. While cardio-respiratory (i.e., aerobic) fitness has been reported to protect against aging-associated cognitive decline, the effects of aerobic fitness on white-matter integrity in aging have remained an open question.

In their recent study, Voss et al. (2012) showed that increased aerobic fitness caused by a one-year walking program predicted enhanced white-matter integrity in prefrontal and temporal lobe areas that are especially susceptible to the adverse effects of aging. Aerobic fitness that resulted from the exercise intervention also predicted short-term memory improvements. These results extend in a very interesting way the scope of applications of diffusion imaging methods and even though the authors failed to see any statistically significant associations between the increased white-matter integrity and short-term memory improvements, these results are nonetheless promising, provide an important contribution to understanding the neuro-cognitive protective effects of aerobic exercise in aging, and suggest that aerobic training is highly important for aging persons. 

ReferenceVoss MW, Heo S, Prakash RS, Erickson KI, Alves H, Chaddock L, Szabo AN, Mailey EL, Wojcicki TR, White SM, Gothe N, McAuley E, Sutton BP, Kramer AF. The influence of aerobic fitness on cerebral white matter integrity and cognitive function in older adults: results of a one-year exercise intervention. Human Brain Mapping (2012), e-publication prior to print. http://dx.doi.org/10.1002/hbm.22119


Beta oscillations decrease in left inferior frontal gyrus when a word violates semantic expectations formed by preceding words

The event-related potential N400 response is elicited when a word violates semantic expectations formed by preceding words in a sentence, for example, the word “river” in sentence “Yesterday, I went to the grocery store to buy a river” would elicit an N400 response. Event-related potentials such as the N400 are obtained by averaging tens or even hundreds of electroencephalogram (EEG) or magnetoencephalogram (MEG) epochs time-locked to onset of stimuli (such as words that violate semantic expectations). The rationale behind this is that the much larger background EEG/MEG activity (that has been presumed to be non-interesting noise) is averaged out, thus leaving the signal of interest for scrutiny. Over the last couple of decades, alternative ways to inspect stimulus-related EEG and MEG activity have been steadily gaining popularity; in one of such approaches, instead of averaging the EEG/MEG trials, power as a function of time and frequency is calculated over the trials, thus allowing inspection of EEG/MEG oscillatory activity that is related to, yet not precisely time-locked to onset of, the stimuli. 

In their recent study, Lin Wang and colleagues (Wang et al. 2012) inspected oscillatory activity elicited by semantic violations that resulted in generation of the N400 event-related response. A correlation was observed between the N400 response (that was predominantly generated in auditory cortical areas) and decreased power in the beta frequency band within left-hemisphere inferior frontal gyrus. The authors suggested that beta suppression reflects larger effort within the task-relevant network (consisting of the left inferior frontal gyrus and that is part of the speech motor system and auditory cortical areas in the left superior temporal region) for attempting to integrate the incongruent word with the preceding context. These findings also add to the pool of evidence indicating that speech motor system plays an important role in speech perception, and further stress the importance of inspecting electromagnetic activity not strictly time-locked to onset of stimuli, in addition to assessing event-related responses, in EEG and MEG studies.

Reference: Wang L, Jensen O, van den Brink D, Weder N, Schoffelen J-M, Magyari L, Hagoort P, Bastiaansen M. Beta oscillations relate to the N400m during language comprehension. Human Brain Mapping (2012), e-publication ahead of print.  http://dx.doi.org/10.1002/hbm.21410


Emotions synchronize brain activity across individuals

Emotions are highly contagious and it is often noted that emotional states can spread rapidly through a large crowd such as when a peaceful demonstration turns into a riot. In their recent study Nummenmaa et al. (2012) shed light on the underlying neural mechanisms. The authors showed short (on average ~90 s) movie clips depicting humans experiencing either strong positive or negative emotions (or in control clips a neutral state) to 16 healthy volunteers during 3-Tesla functional magnetic resonance imaging.  Immediately after the scanning session, the subjects were asked to continuously rate their emotional state while watching the movie clips on two scales: emotional valence (varying from negative to positive emotional state) and emotional arousal (varying from drowsy to highly aroused).

The authors calculated 17-s moving average of inter-subject correlations of brain hemodynamic responses. The resulting inter-subject brain synchrony time courses were then predicted with the emotional valence and arousal ratings to probe which brain areas are synchronized by emotional states. While high arousal predicted across-subjects synchronous brain activity in somatosensory and visual cortices and dorsal attention networks (comprising intraparietal sulci and frontal eye fields), negative valence predicted increased inter-subject synchrony in emotion-processing networks and in the so-called default-mode network (i.e., precuneous, temporoparietal junction, medial prefrontal cortex and posterior superior temporal sulcus). In contrast to negative valence, positive valence was not observed to enhance inter-subject synchrony of brain activity.

Based on these results, the authors proposed that high arousal directs attention of individuals to similar features in their environment, and that negative valence (i.e., unpleasant emotional states) synchronizes the brain areas of individuals that give rise to emotional states and support understanding of the actions of others. The authors further suggested that positive emotions did not result in increased inter-subject synchrony of brain activity due to positive emotions triggering planning of novel and exploratory thoughts and behaviors, whilst negative emotions can be surmised to trigger for example fight-or-flight responses that results in narrowing of mental/behavioral repertoires. The authors conclude that by synchronizing brain activity across individuals, emotions may promote social interaction and facilitate interpersonal understanding.


Nummenmaa L, Glerean E, Viinikainen M, Jaaskelainen IP, Hari R, Sams M. Emotions promote social interaction by synchronizing brain activity across individuals. PNAS (2012) epublication ahead of print.  http://dx.doi.org/10.1073/pnas.1206095109


Emotions are generated by activity patterns of networks of brain areas

The neural basis of emotions is at the same time one of the most interesting and challenging puzzles in cognitive neuroscience. One of the most topical questions is whether specific brain structures generate six basic emotions (fear, happiness, disgust, anger, surprise, sadness) or whether discrete emotions are generated by some other brain mechanisms. A recent theoretical / meta-analysis paper by Lindquist et al. (2012) provides a comprehensive review and insightful synthesis of recent emotion neuroimaging literature. In their meta-analysis, Lindquist et al. specifically focus on addressing the question of whether the particular brain structures that have been implicated to underlie the basic emotions are consistent and specific in their response to a given emotion category; for instance, whether activation of the amygdala is always—and only—related to fearfulness, or whether some other aspect such as stimulus novelty and/or need to quickly determine potential harmfulness of an external stimulus governs amygdala activity (in which case amygdala would contribute rather differently to emotional experiences).

Based on their literature synthesis, Lindquist et al. propose that there is evidence supporting a so-called psychological constructionist approach, which states that discrete emotion categories are constructed by more general activity patterns of networks of brain areas that are not specific to the basic emotion categories. In contrast, little evidence was found supporting the hypothesis that discrete emotion categories would be consistently and specifically localized to distinct brain regions; as one example of this, amygdala that has been traditionally viewed as responding to fear-eliciting stimuli, does respond to stimuli of other emotion categories, habituates quickly, and does not exhibit activity in persons who are anticipating a fear/anxiety-eliciting situation (e.g., stage performance). Importantly, the solid theoretical framework proposed by Lindquist et al. provides multiple empirically testable hypotheses for future neuroimaging studies on the neural basis of emotions.

Reference: Lindquist KA, Wager TD, Kober H, Bliss-Moreau E, Feldman Barrett L. The brain basis of emotion: a meta-analytic review. Behavioral and Brain Sciences (2012) 35: 121-202. http://dx.doi.org/10.1017/S0140525X11000446