This paper appears in the special issue of Psychology and Aging on associative memory. The results show that valuable items are remembered better by younger and older adults, but that memory for contextual details associated with these items may be impaired relative to less valuable items. While older adults often show preserved memory for valuable items. this may come at a cost to memory for incidental detail.
This review article discusses current research in humans and animal models for conditions that favor the shift from actions to habits.
This was a collaborative study from the UCLA Center for Neurocognition and Emotion in Schizophrenia involving labs at the Greater Los Angeles VA, the Semel Institute at UCLA and the Dept. of Psychology. We found impaired recollection for elements of social interactions in patients with schizophrenia or those at high risk of developing the disease. It is very possible that such memory difficulties contribute to problems with social interactions that occur in these patients. See the UCLA press release below
This paper will appear in an upcoming issue of Current Opinion in Behavioral Science on “Habits and Skills”. This paper describes a meta-analysis comparing the locus of activation across a number of different habit learning paradigms. The results suggest that activations from tasks that best fit the criteria for habits are centered around the putamen.
This review paper was a collaborative effort with Michelle Basso in the UCLA Semel Institute. In it, we discuss evidence of an impairment in the integration of memory and sensory information in decision-making in Parkinson’s disease, and that this deficit can lead to the pattern of cognitive deficits displayed by these patients.
This paper resulted from a collaborative effort with the Rissman lab that used diffusion tensor imaging to examine the relationship between memory for high-value items and integrity of tracts related to memory and reward processing. We found that fractional anisotropy of the uncinate fasiculus was related to the number of high value items recalled,. Furthermore, we found that selectivity for high over low value items in memory was related to the number of fibers connecting the nucleus accumbens and ventral tegmental area, implicating reward circuitry.
This paper resulted from the continued collaboration between UCLA and the National Yang-Ming University in Taipei. The goal of this work is to characterize why interleaved practice of motor skills leads to better learning. The results showed that the advantage of interleaved practice was related to greater functional connectivity between the premotor cortex and cerebellum, medial temporal regions, putamen, and thalamus after initial practice. This enhanced connectivity was related to increased functional activation during subsequent training, and may reflect early consolidation of motor skill.
This is another paper from the UCLA Center for Neurocognition and Emotion in Schizophrenia. The main findings were that initial scores on the Wisconsin Card Sorting task and a measure of social cognition were predictive of community functioning in patients with chronic or first-episode schizophrenia. While patients showed a benefit of training on both of these measures, the amount of benefit (learning potential) was not significantly related to community functioning.
This paper follows from Reggente, et al, (2018) which examined the relationship between white matter tracts in the brain and value-directed remembering. In this paper, similar analyses were done in older adults, who exhibit similar levels of memory selectivity as younger adults. In older adults, there was a significant relationship between integrity of the left inferior fronto-occipital fasiculus and how many valuable items were recalled. This relationship was not present for memory for low-value items. This tract may play a role in semantic processing, consistent with the idea that older adults engage in more semantic encoding of valuable items. Unlike in younger adults, selectivity was not associated with integrity of white matter in midbrain reward regions.
This paper is from our collaboration with Alan Castel on the mechanisms of value-directed remembering. Here, subjects study words of different values, with some followed by an instruction to forget the word because it will not be tested, and some followed by an instruction to remember them. We found better memory for high value words, even those that subjects were told to forget, suggesting that value can automatically enhance memory. On the other hand, in this paper we also show that we can substantially diminish effects of value by instructing subjects to use one type of encoding (either deep or shallow) for all of the items. Thus, value directed remembering appears to rely on differential encoding of high value items. In this paper, we make the case that both automatic and strategic effects of value can enhance memory.
This article appears as part of the Research Topics Section on “On the Nature and Scope of Habits and Model-Free Control” edited by John Bargh, Wendy Wood, and David Ellis Melnikoff. Using an avoidance paradigm in which the subject must make responses to avoid unpleasant sounds, we showed that individuals reporting early life adversity exhibiting a greater number of habitual responses than subjects that do not report significant early life adversity. Habitual responses were defined as the subject continuing to make the avoidance response when the earbud delivering the sound was removed- thus, the response occurs through “force of habit” rather than actually having an effect of preventing the sound. This was the first empirical study supported by our NIDA grant to study early life adversity and instrumental learning.
This article appeared in a special issue of Neurobiology of Learning and Memory that featured data presented at the 2018 meeting of the Pavlovian Society. We explored the relationship between early-life stress and habitual responding, finding that childhood physical neglect was a particularly strong predictor of the effect. These data also support the idea that these effects are not primarily mediated by depression, anxiety or socioeconomic status.
Here, we show that older adults under stereotype threat show deficits in their predictions about their memory performance. These results support the idea that exposing older adults to information that highlights age-related cognitive decline results in increased cognitive load that impairs their ability to calibrate their judgments about what they will remember. This study was a collaboration with Alan Castel and appeared in a collection of articles on age-based stereotype threat effects on performance outcomes.
This paper from the Suthana lab addressed mixed findings in the recent literature regarding the effectiveness of medial temporal lobe stimulation in memory enhancement in patients with intracranial electrodes. Based on an analysis of performance across multiple patients and memory tasks, it was found that beneficial effects of stimulation occur when stimulating electrodes are placed in the white matter of entorhinal cortex, suggesting that activation of the perforant path input to the hippocampus may be beneficial for memory encoding.
This paper demonstrates that the benefits of interleaved practice extend to implicit learning of fine-motor sequences. Participants performed a serial reaction time task in which key press locations occurred according to three different sequences. Participants were not aware that the locations were not random. When tested after a delay, participants were faster if they practiced interleaved sequences rather than blocked sequences. Interleaved practice also led to better performance when new sequences were learned. These results suggest that explicit retrieval is not necessary for the benefit of interleaved practice.
In this collaboration with Michele Basso and Jochen Ditterich we showed that base-rate priors can influence perceptual decision-making even when the participant is not aware of them. Participants judged the orientation of red and green dot patterns that varied in terms of coherence. Unbeknownst to the participants, the different colors were associated with more frequent left or right tilted stimuli. Participants incorporated this information into their decisions as shown by a bias when judging low coherence stimuli. Using the Drift-Diffusion model, we found that color-specific biases were implemented through drift rate changes.
This review written with Alan Castel outlines a framework in which valuable or important information is prioritized at encoding by both automatic mechanisms and strategic differential application of semantic encoding processes.