Bradley Monk

Humans have a unique cognitive ability known as relational abstraction, which allows us to identify logical rules and patterns beyond basic perceptual characteristics. This ability represents a key difference between how humans and other animals learn and interact with the world. With current large language models rivaling human intelligence in many regards, we investigated whether relational abstraction was an emergent chat completion capacity in various models. We find that despite their impressive language processing skills, all tested language models failed the relational match-to-sample (RMTS) test, a benchmark for assessing relational abstraction. These results challenge the assumption that advanced language skills inherently confer the capacity for complex relational reasoning. The paper highlights the need for a broader evaluation of AI cognitive abilities, emphasizing that language proficiency alone may not be indicative of certain higher-order cognitive processes thought to be supported by language.

PURPOSE: Balstilimab (antiprogrammed death-1) and zalifrelimab (anticytotoxic T-lymphocyte–associated antigen-4) are two new checkpoint inhibitors emerging as promising investigational agents for the treatment of advanced cervical cancer. This phase II trial (ClinicalTrials.gov identifier: NCT03495882 ) evaluated the combination of balstilimab plus zalifrelimab in patients with recurrent and/or metastatic cervical cancer who relapsed after prior platinum-based therapy. PATIENTS AND METHODS: Patients were intravenously dosed with balstilimab 3 mg/kg once every 2 weeks and zalifrelimab 1 mg/kg once every 6 weeks, for up to 24 months. The primary end point was objective response rate (ORR, RECIST version 1.1, assessed by independent central review). Secondary end points included duration of response, safety and tolerability, and survival. RESULTS: In total, 155 women (median age, 50 years [range, 24-76 years]) were enrolled and treated with balstilimab plus zalifrelimab; 125 patients had measurable disease at baseline and one prior line of platinum-based therapy in the advanced setting, and these patients constituted the efficacy-evaluable population. The median follow-up was 21 months. The confirmed ORR was 25.6% (95% CI, 18.8 to 33.9), including 10 complete responders and 22 partial responders, with median duration of response not reached (86.5%, 75.5%, and 64.2% at 6, 9, and 12 months, respectively). The ORRs were 32.8% and 9.1% in patients with programmed death ligand-1–positive and programmed death ligand-1–negative tumors, respectively. For patients with squamous cell carcinoma, the ORR was 32.6%. The overall disease control rate was 52% (95% CI, 43.3 to 60.6). Hypothyroidism (14.2%) and hyperthyroidism (7.1%) were the most common immune-mediated adverse events. CONCLUSION: Promising and durable clinical activity, with favorable tolerability, was seen in this largest trial to date evaluating dual programmed death-1/cytotoxic T-lymphocyte–associated antigen-4 blockade in patients with recurrent and/or metastatic cervical cancer. Further investigation of the balstilimab and zalifrelimab combination in this setting is continuing.

There is hope that genomic information will assist prediction, treatment, and understanding of Alzheimer’s disease (AD). Here, using exome data from ∼10,000 individuals, we explore machine learning neural network (NN) methods to estimate the impact of SNPs (i.e., genetic variants) on AD risk. We develop an NN-based method (netSNP) that identifies hundreds of novel potentially protective or at-risk AD-associated SNPs (along with an effect measure); the majority with frequency under 0.01. For case individuals, the number of “protective” (or “at-risk”) netSNP-identified SNPs in their genome correlates positively (or inversely) with their age of AD diagnosis and inversely (or positively) with autopsy neuropathology. The effect measure increases correlations. Simulations suggest our results are not due to genetic linkage, overfitting, or bias introduced by netSNP. These findings suggest that netSNP can identify SNPs associated with AD pathophysiology that may assist with the diagnosis and mechanistic understanding of the disease.

Misophonia is a newly researched condition in which specific sounds cause an intense, aversive response in individuals, characterized by negative emotions and autonomic arousal. Although virtually any sound can become a misophonic “trigger,” the most common sounds appear to be bodily sounds related to chewing and eating as well as other repetitive sounds. An intriguing aspect of misophonia is the fact that many misophonic individuals report that they are triggered more, or even only, by sounds produced by specific individuals, and less, or not at all, by sounds produced by animals (although there are always exceptions). In general, anecdotal evidence suggests that misophonic triggers involve a combination of sound stimuli and contextual cues. The aversive stimulus is more than just a sound and can be thought of as a Gestalt of features which includes sound as a necessary component as well as additional contextual information. In this study, we explore how contextual information influences misophonic responses to human chewing, as well as sonically similar sounds produced by non-human sources. The current study revealed that the exact same sound can be perceived as being much more or less aversive depending on the contextual information presented alongside the auditory information. The results of this study provide a foundation for potential cognitive based therapies.

Neuronal activity in the lateral habenula (LHb), a brain region implicated in depression [C. D. Proulx, O. Hikosaka, R. Malinow,Nat. Neurosci.17, 1146–1152 (2014)], decreases during reward and increases during punishment or reward omission [M. Matsumoto, O. Hikosaka,Nature447, 1111–1115 (2007)]. While stress is a major risk factor for depression and strongly impacts the LHb, its effect on LHb reward signals is unknown. Here we image LHb neuronal activity in behaving mice and find that acute stress transforms LHb reward responses into punishment-like neural signals; punishment-like responses to reward omission also increase. These neural changes matched the onset of anhedonic behavior and were specific to LHb neurons that distinguished reward and its omission. Thus, stress distorts LHb responsivity to positive and negative feedback, which could bias individuals toward negative expectations, a key aspect of the proposed pathogenesis of depression [A. T. Beck,Depression: Clinical, Experimental, and Theoretical Aspects, sixth Ed (1967)].

Significance: The lateral habenula, a brain region that has been implicated in depression, receives inputs from brain nuclei associated with basic emotions and drives. In this report, using fiber photometry and optogenetics on behaving rats, we show that one major lateral habenula output pathway controls the motivation to exert effort in both aversive and appetitive contexts. Overactivity of this pathway could contribute to the reduced motivation seen in human depression.

The atypical PKC isoforms, PKMζ and PKCλ have been proposed as integral substrates of long-term memory. Inhibition of these isoforms has recently been demonstrated to be sufficient for impairing the expression and maintenance of long-term potentiation. Additionally, the pseudosubstrate inhibitor, zeta inhibitory peptide (ZIP), which effectively blocks PKMζ and PKCλ, has previously been shown to disrupt associative memory; very little is known about its effects on pathological nonassociative forms of memory related to addiction. The neural and molecular substrates of memory and addiction have recently been argued to overlap. Here, we used ZIP to disrupt PKMζ and PKCλ activity to examine their role in cocaine sensitization, a nonassociative, addiction-related memory argued to underlie the transition from casual to pathological drug use. We examined the effects of both continuous and acute administration of ZIP. Even a single application of ZIP blocked the development of sensitization; sustained inhibition using osmotic pumps produced an almost complete blockade of sensitization. Further, a single application of ZIP was shown to reduce membrane-bound AMPAR expression. These results demonstrate a novel, critical role for the atypical PKC isoforms in nonassociative memory and cocaine addiction.

Prenatal alcohol exposure leads to long‐lasting cognitive and attention deficits, as well as hyperactivity. Using a rat model, we have previously shown that perinatal supplementation with the essential nutrient, choline, can reduce the severity of some fetal alcohol effects, including hyperactivity and deficits in learning and memory. In fact, choline can mitigate alcohol‐related learning deficits even when administered after developmental alcohol exposure, during the postnatal period. However, it is not yet known how choline is able to mitigate alcohol‐related behavioral alterations. Choline may act by altering cholinergic signaling in the hippocampus. This study examined the effects of developmental alcohol exposure and perinatal choline supplementation on hippocampal M₁and M_2/4muscarinic receptors. Sprague‐Dawley rat pups were orally intubated with ethanol (5.25 mg/kg/day) from postnatal days (PD) 4–9, a period of brain development equivalent to the human third trimester; control subjects received sham intubations. From PD 4–30, subjects were injected s.c. with choline chloride (100 mg/kg/day) or saline vehicle. Open field activity was assessed from PD 30 through 33, and brain tissue was collected on PD 35 for autoradiographic analysis. Ethanol‐exposed subjects were more active compared to controls during the first 2 days of testing, an effect attenuated with choline supplementation. Developmental alcohol exposure significantly decreased the density of muscarinic M₁receptors in the dorsal hippocampus, an effect that was not altered by choline supplementation. In contrast, developmental alcohol exposure significantly increased M_2/4receptor density, an effect mitigated by choline supplementation. In fact, M_2/4receptor density of subjects exposed to alcohol and treated with choline did not differ significantly from that of controls. These data suggest that developmental alcohol exposure can cause long‐lasting changes in the hippocampal cholinergic system and that perinatal choline supplementation may attenuate alcohol‐related behavioral changes by influencing cholinergic systems.

BACKGROUND: Prenatal alcohol exposure can alter physical and behavioral development, leading to a range of fetal alcohol spectrum disorders. Despite warning labels, pregnant women continue to drink alcohol, creating a need to identify effective interventions to reduce the severity of alcohol's teratogenic effects. Choline is an essential nutrient that influences brain and behavioral development. Recent studies indicate that choline supplementation can reduce the teratogenic effects of developmental alcohol exposure. The present study examined whether choline supplementation during prenatal ethanol treatment could mitigate the adverse effects of ethanol on behavioral development.METHODS: Pregnant Sprague‐Dawley rats were intubated with 6 g/kg/day ethanol in a binge‐like manner from gestational days 5–20; pair‐fed and ad libitum chow controls were included. During treatment, subjects from each group were intubated with either 250 mg/kg/day choline chloride or vehicle. Spontaneous alternation, parallel bar motor coordination, Morris water maze, and spatial working memory were assessed in male and female offspring.RESULTS: Subjects prenatally exposed to alcohol exhibited delayed development of spontaneous alternation behavior and deficits on the working memory version of the Morris water maze during adulthood, effects that were mitigated with prenatal choline supplementation. Neither alcohol nor choline influenced performance on the motor coordination task.CONCLUSIONS: These data indicate that choline supplementation during prenatal alcohol exposure may reduce the severity of fetal alcohol effects, particularly on alterations in tasks that require behavioral flexibility. These findings have important implications for children of women who drink alcohol during pregnancy. Birth Defects Research (Part A), 2010.