Anthony R. McIntosh

Aging Gracefully: Compensatory Brain Activity in High-Performing Older Adults

Whereas some older adults show significant cognitive deficits, others perform as well as young adults. We investigated the neural basis of these different aging patterns using positron emission tomography (PET). In PET and functional MRI (fMRI) studies, prefrontal cortex (PFC) activity tends to be less asymmetric in older than in younger adults (Hemispheric Asymmetry Reduction in Old Adults or HAROLD). This change may help counteract age-related neurocognitive decline (compensation hypothesis) or it may reflect an age-related difficulty in recruiting specialized neural mechanisms (dedifferentiation hypothesis). To compare these two hypotheses, we measured PFC activity in younger adults, low-performing older adults, and high-performing older adults during recall and source memory of recently studied words. Compared to recall, source memory was associated with right PFC activations in younger adults. Low-performing older adults recruited similar right PFC regions as young adults, but high-performing older adults engaged PFC regions bilaterally. Thus, consistent with the compensation hypothesis and inconsistent with the dedifferentiation hypothesis, a hemispheric asymmetry reduction was found in high-performing but not in low-performing older adults. The results suggest that low-performing older adults recruited a similar network as young adults but used it inefficiently, whereas high-performing older adults counteracted age-related neural decline through a plastic reorganization of neurocognitive networks.

Spatial Pattern Analysis of Functional Brain Images Using Partial Least Squares

This paper introduces a new tool for functional neuroimage analysis: partial least squares (PLS). It is unique as a multivariate method in its choice of emphasis for analysis, that being the covariance between brain images and exogenous blocks representing either the experiment design or some behavioral measure. What emerges are spatial patterns of brain activity that represent the optimal association between the images and either of the blocks. This process differs substantially from other multivariate methods in that rather than attempting to predict the individual values of the image pixels, PLS attempts to explain the relation between image pixels and task or behavior. Data from a face encoding and recognition PET rCBF study are used to illustrate two types of PLS analysis: an activation analysis of task with images and a brain-behavior analysis. The commonalities across the two analyses are suggestive of a general face memory network differentially engaged during encoding and recognition. PLS thus serves as an important extension by extracting new information from imaging data that is not accessible through other currently used univariate and multivariate image analysis tools.

Emerging concepts for the dynamical organization of resting-state activity in the brain

A broad body of experimental work has demonstrated that apparently spontaneous brain activity is not random. At the level of large-scale neural systems, as measured with functional MRI (fMRI), this ongoing activity reflects the organization of a series of highly coherent functional networks. These so-called resting-state networks (RSNs) closely relate to the underlying anatomical connectivity but cannot be understood in those terms alone. Here we review three large-scale neural system models of primate neocortex that emphasize the key contributions of local dynamics, signal transmission delays and noise to the emerging RSNs. We propose that the formation and dissolution of resting-state patterns reflects the exploration of possible functional network configurations around a stable anatomical skeleton.

Age-Related Differences in Neural Activity during Memory Encoding and Retrieval: A Positron Emission Tomography Study

Positron emission tomography (PET) was used to compare regional cerebral blood flow (rCBF) in young (mean 26 years) and old (mean 70 years) subjects while they were encoding, recognizing, and recalling word pairs. A multivariate partial-least-squares (PLS) analysis of the data was used to identify age-related neural changes associated with (1) encoding versus retrieval and (2) recognition versus recall. Young subjects showed higher activation than old subjects (1) in left prefrontal and occipito-temporal regions during encoding and (2) in right prefrontal and parietal regions during retrieval. Old subjects showed relatively higher activation than young subjects in several regions, including insular regions during encoding, cuneus/precuneus regions during recognition, and left prefrontal regions during recall. Frontal activity in young subjects was left-lateralized during encoding and right-lateralized during recall [hemispheric encoding/retrieval asymmetry (HERA)], whereas old adults showed little frontal activity during encoding and a more bilateral pattern of frontal activation during retrieval. In young subjects, activation in recall was higher than that in recognition in cerebellar and cingulate regions, whereas recognition showed higher activity in right temporal and parietal regions. In old subjects, the differences in blood flow between recall and recognition were smaller in these regions, yet more pronounced in other regions. Taken together, the results indicate that advanced age is associated with neural changes in the brain systems underlying encoding, recognition, and recall. These changes take two forms: (1) age-related decreases in local regional activity, which may signal less efficient processing by the old, and (2) age-related increases in activity, which may signal functional compensation.

Partial least squares analysis of neuroimaging data: applications and advances.

Partial least squares (PLS) analysis has been used to characterize distributed signals measured by neuroimaging methods like positron emission tomography (PET), functional magnetic resonance imaging (fMRI), event-related potentials (ERP) and magnetoencephalography (MEG). In the application to PET, it has been used to extract activity patterns differentiating cognitive tasks, patterns relating distributed activity to behavior, and to describe large-scale interregional interactions or functional connections. This paper reviews the more recent extension of PLS to the analysis of spatiotemporal patterns present in fMRI, ERP, and MEG data. We present a basic mathematical description of PLS and discuss the statistical assessment using permutation testing and bootstrap resampling. These two resampling methods provide complementary information of the statistical strength of the extracted activity patterns (permutation test) and the reliability of regional contributions to the patterns (bootstrap resampling). Simulated ERP data are used to guide the basic interpretation of spatiotemporal PLS results, and examples from empirical ERP and fMRI data sets are used for further illustration. We conclude with a discussion of some caveats in the use of PLS, including nonlinearities, nonorthogonality, and interpretation difficulties. We further discuss its role as an important tool in a pluralistic analytic approach to neuroimaging.

Partial Least Squares (PLS) methods for neuroimaging: a tutorial and review.

Partial Least Squares (PLS) methods are particularly suited to the analysis of relationships between measures of brain activity and of behavior or experimental design. In neuroimaging, PLS refers to two related methods: (1) symmetric PLS or Partial Least Squares Correlation (PLSC), and (2) asymmetric PLS or Partial Least Squares Regression (PLSR). The most popular (by far) version of PLS for neuroimaging is PLSC. It exists in several varieties based on the type of data that are related to brain activity: behavior PLSC analyzes the relationship between brain activity and behavioral data, task PLSC analyzes how brain activity relates to pre-defined categories or experimental design, seed PLSC analyzes the pattern of connectivity between brain regions, and multi-block or multi-table PLSC integrates one or more of these varieties in a common analysis. PLSR, in contrast to PLSC, is a predictive technique which, typically, predicts behavior (or design) from brain activity. For both PLS methods, statistical inferences are implemented using cross-validation techniques to identify significant patterns of voxel activation. This paper presents both PLS methods and illustrates them with small numerical examples and typical applications in neuroimaging.

Key role of coupling, delay, and noise in resting brain fluctuations

A growing body of neuroimaging research has documented that, in the absence of an explicit task, the brain shows temporally coherent activity. This so-called “resting state” activity or, more explicitly, the default-mode network, has been associated with daydreaming, free association, stream of consciousness, or inner rehearsal in humans, but similar patterns have also been found under anesthesia and in monkeys. Spatiotemporal activity patterns in the default-mode network are both complex and consistent, which raises the question whether they are the expression of an interesting cognitive architecture or the consequence of intrinsic network constraints. In numerical simulation, we studied the dynamics of a simplified cortical network using 38 noise-driven (Wilson–Cowan) oscillators, which in isolation remain just below their oscillatory threshold. Time delay coupling based on lengths and strengths of primate corticocortical pathways leads to the emergence of 2 sets of 40-Hz oscillators. The sets showed synchronization that was anticorrelated at <0.1 Hz across the sets in line with a wide range of recent experimental observations. Systematic variation of conduction velocity, coupling strength, and noise level indicate a high sensitivity of emerging synchrony as well as simulated blood flow blood oxygen level-dependent (BOLD) on the underlying parameter values. Optimal sensitivity was observed around conduction velocities of 1–2 m/s, with very weak coupling between oscillators. An additional finding was that the optimal noise level had a characteristic scale, indicating the presence of stochastic resonance, which allows the network dynamics to respond with high sensitivity to changes in diffuse feedback activity.

Noise during rest enables the exploration of the brain's dynamic repertoire.

Traditionally brain function is studied through measuring physiological responses in controlled sensory, motor, and cognitive paradigms. However, even at rest, in the absence of overt goal-directed behavior, collections of cortical regions consistently show temporally coherent activity. In humans, these resting state networks have been shown to greatly overlap with functional architectures present during consciously directed activity, which motivates the interpretation of rest activity as day dreaming, free association, stream of consciousness, and inner rehearsal. In monkeys, it has been shown though that similar coherent fluctuations are present during deep anesthesia when there is no consciousness. Here, we show that comparable resting state networks emerge from a stability analysis of the network dynamics using biologically realistic primate brain connectivity, although anatomical information alone does not identify the network. We specifically demonstrate that noise and time delays via propagation along connecting fibres are essential for the emergence of the coherent fluctuations of the default network. The spatiotemporal network dynamics evolves on multiple temporal scales and displays the intermittent neuroelectric oscillations in the fast frequency regimes, 1–100 Hz, commonly observed in electroencephalographic and magnetoencephalographic recordings, as well as the hemodynamic oscillations in the ultraslow regimes, <0.1 Hz, observed in functional magnetic resonance imaging. The combination of anatomical structure and time delays creates a space–time structure in which the neural noise enables the brain to explore various functional configurations representing its dynamic repertoire.

Functional brain maps of retrieval mode and recovery of episodic information.

Positron emission tomography (PET) was used to identify brain regions associated with two component processes of episodic retrieval; those related to thinking back in subjective time (retrieval mode) and those related to actual recovery of stored information (ecphory). Healthy young subjects recognized words that had been encoded with respect to meaning or the speakers voice. Regardless of how the information had been encoded, recognition was associated with increased activation in regions in right prefrontal cortex, left anterior cingulate, and cerebellum. These activations reflect retrieval mode. Recognition following meaning encoding was specifically associated with increased activation in left temporal cortex, and recognition following voice encoding involved regions in right orbital frontal and parahippocampal cortex. These activations reflect ecphory of differentially encoded information.

Interactions Between the Nucleus Accumbens and Auditory Cortices Predict Music Reward Value

Music Was My First Love Why do human beings enjoy music? Salimpoor et al. (p. 216) combined behavioral economics with brain scanning to explore how a piece of music is considered rewarding to an individual when it is heard for the first time. They discovered that neural activity in the mesolimbic striatum during listening to a novel piece of music was the best predictor of the money listeners were willing to spend on buying the piece. These observations implicate sensory cortical areas in reward processing, which the authors attribute to the aesthetic nature of the judgment. Activity in the mesolimbic striatum and its interaction with auditory cortices determines the hedonic response to new music. We used functional magnetic resonance imaging to investigate neural processes when music gains reward value the first time it is heard. The degree of activity in the mesolimbic striatal regions, especially the nucleus accumbens, during music listening was the best predictor of the amount listeners were willing to spend on previously unheard music in an auction paradigm. Importantly, the auditory cortices, amygdala, and ventromedial prefrontal regions showed increased activity during listening conditions requiring valuation, but did not predict reward value, which was instead predicted by increasing functional connectivity of these regions with the nucleus accumbens as the reward value increased. Thus, aesthetic rewards arise from the interaction between mesolimbic reward circuitry and cortical networks involved in perceptual analysis and valuation.