Search results
Found 12849 matches for
The Alignment Problem Machine Learning and Human Values
When the systems we attempt to teach will not, in the end, do what we want or what we expect, ethical and potentially existential risks emerge. Researchers call this the alignment problem.
The Most Human Human What Artificial Intelligence Teaches Us About Being Alive
Art and science meet an engaged mind and the friction produces real fire.” —The New Yorker Each year, the AI community convenes to administer the famous (and famously controversial) Turing test, pitting sophisticated software programs ...
Algorithms to Live By The Computer Science of Human Decisions
From finding a spouse to finding a parking spot, from organizing one’s inbox to peering into the future, Algorithms to Live By transforms the wisdom of computer science into strategies for human living.
Using category structures to test iterated learning as a method for identifying inductive biases.
Many of the problems studied in cognitive science are inductive problems, requiring people to evaluate hypotheses in the light of data. The key to solving these problems successfully is having the right inductive biases-assumptions about the world that make it possible to choose between hypotheses that are equally consistent with the observed data. This article explores a novel experimental method for identifying the biases that guide human inductive inferences. The idea behind this method is simple: This article uses the responses produced by a participant on one trial to generate the stimuli that either they or another participant will see on the next. A formal analysis of this "iterated learning" procedure, based on the assumption that the learners are Bayesian agents, predicts that it should reveal the inductive biases of these learners, as expressed in a prior probability distribution over hypotheses. This article presents a series of experiments using stimuli based on a well-studied set of category structures, demonstrating that iterated learning can be used to reveal the inductive biases of human learners.
Validation of Korean Version of the Oxford Cognitive Screen (K-OCS), a Post Stroke-Specific Cognitive Screening Tool.
OBJECTIVE: To establish and evaluate the validity of the recently developed Korean version of the Oxford Cognitive Screen (K-OCS), this study verified its reliability, validity, and diagnostic accuracy. METHODS: Between November 2021 and December 2023, we recruited 72 patients with stroke from our hospital who agreed to participate in the study. The patients were repeatedly tested using K-OCS by the same or different assessors to estimate inter- and intra-rater reliability. To demonstrate the validity and usability of K-OCS, the test results of screening tools currently used in clinical practice, including the Korean-Mini Mental State Examination and the Korean version of the Montreal Cognitive Assessment, were used in comparison analyses. RESULTS: The subtests of K-OCS demonstrated excellent inter-rater reliability (intra-class correlation coefficient [ICC]=0.914-0.998) and test-retest reliability (ICC=0.913-0.994). We found moderate-to-strong correlations for convergent validity for the subsets (r=0.378- 0.979, p<0.01), and low-to-moderate discriminant validity correlations. The optimal cut-offs estimated for the subtests of the K-OCS showed a good-to-high range of specificity (94.8%- 100%). The positive predictive value was 58.2%-100% and negative predictive value was 65.6%-98.4%. Sensitivity was estimated at 25.6%-86.9%. CONCLUSION: The results of this study indicate that K-OCS is a reliable and valid tool for screening cognitive impairment in patients post-stroke.
Psychological booster shots targeting memory increase long-term resistance against misinformation
Abstract An increasing number of real-world interventions aim to preemptively protect or inoculate people against misinformation. Inoculation research has demonstrated positive effects on misinformation resilience when measured immediately after treatment via messages, games, or videos. However, very little is currently known about their long-term effectiveness and the mechanisms by which such treatment effects decay over time. We start by proposing three possible models on the mechanisms driving resistance to misinformation. We then report five pre-registered longitudinal experiments (N total = 11,759) that investigate the effectiveness of psychological inoculation interventions over time as well as their underlying mechanisms. We find that text-based and video-based inoculation interventions can remain effective for one month—whereas game-based interventions appear to decay more rapidly—and that memory-enhancing booster interventions can enhance the diminishing effects of counter-misinformation interventions. Finally, we propose an integrated memory-motivation model, concluding that misinformation researchers would benefit from integrating knowledge from the cognitive science of memory to design better psychological interventions that can counter misinformation durably over time and at-scale.
Examining two of the ingredients of Cognitive therapy for adolescent social anxiety disorder: Back-translation from a treatment trial.
BACKGROUND: Cognitive Therapy for Social Anxiety Disorder (CT-SAD) based on the Clark & Wells model is a complex intervention comprised of a series of therapeutic elements. Two of the key ingredients are the self-focused attention and safety behaviour experiment and video feedback. The present study examined the effects of these two therapeutic procedures in adolescents with SAD, as well as common themes of the young people's social fears and negative self-images. METHOD: 35 participants with a diagnosis of SAD completed internet-delivered CT-SAD as part of a randomised controlled trial. We conducted a series of paired samples t-tests to evaluate the effects of the self-focused attention and safety behaviour experiment and video feedback. We applied Latent Dirichlet Allocation to identify latent topics based on participants' description of their social fears and negative self-images that were elicited during the course of these therapy procedures. RESULTS: Participants reported lower anxiety and more positive self-appraisals when focusing externally and dropping safety behaviours, compared to when focusing internally and using safety behaviours (ps
Fronto-striatal organization: Defining functional and microstructural substrates of behavioural flexibility
Discrete yet overlapping frontal-striatal circuits mediate broadly dissociable cognitive and behavioural processes. Using a recently developed multi-echo resting-state functional MRI (magnetic resonance imaging) sequence with greatly enhanced signal compared to noise ratios, we map frontal cortical functional projections to the striatum and striatal projections through the direct and indirect basal ganglia circuit. We demonstrate distinct limbic (ventromedial prefrontal regions, ventral striatum - VS, ventral tegmental area - VTA), motor (supplementary motor areas - SMAs, putamen, substantia nigra) and cognitive (lateral prefrontal and caudate) functional connectivity. We confirm the functional nature of the cortico-striatal connections, demonstrating correlates of well-established goal-directed behaviour (involving medial orbitofrontal cortex - mOFC and VS), probabilistic reversal learning (lateral orbitofrontal cortex - lOFC and VS) and attentional shifting (dorsolateral prefrontal cortex - dlPFC and VS) while assessing habitual model-free (SMA and putamen) behaviours on an exploratory basis. We further use neurite orientation dispersion and density imaging (NODDI) to show that more goal-directed model-based learning (MBc) is also associated with higher mOFC neurite density and habitual model-free learning (MFc) implicates neurite complexity in the putamen. This data highlights similarities between a computational account of MFc and conventional measures of habit learning. We highlight the intrinsic functional and structural architecture of parallel systems of behavioural control.
Novelty, conditioning and attentional bias to sexual rewards
The Internet provides a large source of novel and rewarding stimuli, particularly with respect to sexually explicit materials. Novelty-seeking and cue-conditioning are fundamental processes underlying preference and approach behaviors implicated in disorders of addiction. Here we examine these processes in individuals with compulsive sexual behaviors (CSB), hypothesizing a greater preference for sexual novelty and stimuli conditioned to sexual rewards relative to healthy volunteers. Twenty-two CSB males and forty age-matched male volunteers were tested in two separate behavioral tasks focusing on preferences for novelty and conditioned stimuli. Twenty subjects from each group were also assessed in a third conditioning and extinction task using functional magnetic resonance imaging. CSB was associated with enhanced novelty preference for sexual, as compared to control images, and a generalized preference for cues conditioned to sexual and monetary versus neutral outcomes compared to healthy volunteers. CSB individuals also had greater dorsal cingulate habituation to repeated sexual versus monetary images with the degree of habituation correlating with enhanced preference for sexual novelty. Approach behaviors to sexually conditioned cues dissociable from novelty preference were associated with an early attentional bias to sexual images. This study shows that CSB individuals have a dysfunctional enhanced preference for sexual novelty possibly mediated by greater cingulate habituation along with a generalized enhancement of conditioning to rewards. We further emphasize a dissociable role for cue-conditioning and novelty preference on the early attentional bias for sexual cues. These findings have wider relevance as the Internet provides a broad range of novel and potentially rewarding stimuli.