2022

Language models show human-like content effects on reasoning tasks

Dasgupta, Ishita, Lampinen, Andrew K., Chan, Stephanie C. Y. et al.

Understand

Reasoning is a key ability for an intelligent system.

Large language models (LMs) achieve above-chance performance on abstract reasoning tasks, but exhibit many imperfections.
However, human abstract reasoning is also imperfect.
For example, human reasoning is affected by our real-world knowledge and beliefs, and shows notable "content effects"; humans reason more reliably when the semantic content of a problem supports the correct logical inferences.

Built on

Journal of Experimental Psychology
C. R. Peterson, Z. Ulehla, Uncertainty, inference difficulty, and probability learning · 1964
Earlier work this paper cites.
Quarterly Journal of Experimental Psychology
P. C. Wason, Reasoning about a rule · 1968
Earlier work this paper cites.
Quarterly journal of experimental psychology
P. C. Wason, D. Shapiro, Natural and contrived experience in a reasoning problem · 1971
Earlier work this paper cites.
International Journal of Psychology
A. K. Luria, Towards the problem of the historical nature of psychological processes · 1971
Earlier work this paper cites.
British journal of Psychology
P. N. Johnson-Laird, P. Legrenzi, M. S. Legrenzi, Reasoning and a sense of reality · 1972
Earlier work this paper cites.
P. C. Wason, P. N. Johnson-Laird, Psychology of reasoning: Structure and content
1972
Earlier work this paper cites.
Acta psychologica
W. A. Wickelgren, Speed-accuracy tradeoff and information processing dynamics · 1977
Earlier work this paper cites.
Cognitive science
A. Newell, Physical symbol systems · 1980
Earlier work this paper cites.
D. Kahneman, S. P. Slovic, P. Slovic, A. Tversky, Judgment under uncertainty: Heuristics and biases
1982
Earlier work this paper cites.
Memory & cognition
J. Evans, J. L. Barston, P. Pollard, On the conflict between logic and belief in syllogistic reasoning · 1983
Earlier work this paper cites.
British Journal of Psychology
J. S. B. Evans, Heuristic and analytic processes in reasoning · 1984
Earlier work this paper cites.
Cognitive psychology
P. W. Cheng, K. J. Holyoak, Pragmatic reasoning schemas · 1985
Earlier work this paper cites.
Cognition
J. A. Fodor, Z. W. Pylyshyn, Connectionism and cognitive architecture: A critical analysis · 1988
Earlier work this paper cites.
Cognition
L. Cosmides, The logic of social exchange: Has natural selection shaped how humans reason? studies with the wason selection task · 1989
Earlier work this paper cites.
J. S. B. Evans, Bias in human reasoning: Causes and consequences
1989
Earlier work this paper cites.
Psychological review
J. D. Cohen, K. Dunbar, J. L. McClelland, On the control of automatic processes: a parallel distributed processing account of the stroop effect · 1990
Earlier work this paper cites.
H. A. Simon, Utility and probability
1990
Earlier work this paper cites.
Developmental Psychology
D. R. Lehman, R. E. Nisbett, A longitudinal study of the effects of undergraduate training on reasoning · 1990
Earlier work this paper cites.
The adapted mind: Evolutionary psychology and the generation of culture
L. Cosmides, J. Tooby, Cognitive adaptations for social exchange · 1992
Earlier work this paper cites.
European Journal of Cognitive Psychology
J. S. B. Evans, S. Newstead, J. Allen, P. Pollard, Debiasing by instruction: The case of belief bias · 1994
Earlier work this paper cites.
Cahiers de Psychologie Cognitive/Current Psychology of Cognition
J. S. B. Evans, T. S. Perry, Belief bias in children’s reasoning · 1995
Earlier work this paper cites.
H. H. Clark, Using language
1996
Earlier work this paper cites.
Journal of Memory and Language
Original
J. S. B. Evans, J. Clibbens, B. Rood, The role of implicit and explicit negation in conditional reasoning bias · 1996
Earlier work this paper cites.
Annual review of psychology
P. N. Johnson-Laird, Deductive reasoning · 1999
Earlier work this paper cites.
Psychological review
K. C. Klauer, J. Musch, B. Naumer, On belief bias in syllogistic reasoning · 2000
Earlier work this paper cites.
G. F. Marcus, The algebraic mind: Integrating connectionism and cognitive science
2003
Earlier work this paper cites.
Trends in cognitive sciences
J. S. B. Evans, In two minds: dual-process accounts of reasoning · 2003
Earlier work this paper cites.
Psychonomic Bulletin & Review
M. Oaksford, N. Chater, Optimal data selection: Revision, review, and reevaluation · 2003
Earlier work this paper cites.
International Group for the Psychology of Mathematics Education
M. Inglis, A. Simpson, Mathematicians and the selection task · 2004
Earlier work this paper cites.
Thinking & Reasoning
J. S. B. Evans, J. Curtis-Holmes, Rapid responding increases belief bias: Evidence for the dual-process theory of reasoning · 2005
Earlier work this paper cites.
Applied Measurement in Education
S. L. Wise, X. Kong, Response time effort: A new measure of examinee motivation in computer-based tests · 2005
Earlier work this paper cites.
B. MacCartney, C. D. Manning, Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing
2007
Earlier work this paper cites.
G. Marcus, Kluge: The haphazard evolution of the human mind
2009
Earlier work this paper cites.
Trends in cognitive sciences
J. L. McClelland, M. M. Botvinick, D. C. Noelle, D. C. Plaut, T. T. Rogers, M. S. Seidenberg, L. B. Smith, Letting structure emerge: connectionist and dynamical systems approaches to cognition · 2010
Earlier work this paper cites.
Psychological review
C. Dube, C. M. Rotello, E. Heit, Assessing the belief bias effect with rocs: it’s a response bias effect · 2010
Earlier work this paper cites.
Annual review of psychology
G. Gigerenzer, W. Gaissmaier, Heuristic decision making · 2011
Earlier work this paper cites.
D. Kahneman, Thinking, fast and slow
2011
Earlier work this paper cites.
J. S. B. Evans, D. E. Over, Rationality and reasoning
2013
Earlier work this paper cites.
Cognition
D. Trippas, M. F. Verde, S. J. Handley, Using forced choice to test belief bias in syllogistic reasoning · 2014
Earlier work this paper cites.
Cognitive science
M. M. Botvinick, J. D. Cohen, The computational and neural basis of cognitive control: charted territory and new frontiers · 2014
Earlier work this paper cites.

Similar

Proceedings of the national academy of sciences
D. L. Yamins, H. Hong, C. F. Cadieu, E. A. Solomon, D. Seibert, J. J. DiCarlo, Performance-optimized hierarchical models predict neural responses in higher visual cortex · 2014
Cited alongside, same era.
The Routledge handbook of embodied cognition
M. W. Alibali, R. Boncoddo, A. B. Hostetter, Gesture in reasoning: An embodied perspective · 2014
Cited alongside, same era.
arXiv preprint arXiv:1508.05326
Original
S. R. Bowman, G. Angeli, C. Potts, C. D. Manning, A large annotated corpus for learning natural language inference · 2015
Cited alongside, same era.
Proceedings of ESSLLI
M. H. Tessler, Understanding belief bias by measuring prior beliefs for a bayesian model of syllogistic reasoning · 2015
Cited alongside, same era.
Nature neuroscience
D. L. Yamins, J. J. DiCarlo, Using goal-driven deep learning models to understand sensory cortex · 2016
Cited alongside, same era.
N. Attridge, A. Aberdein, M. Inglis, Does studying logic improve logical reasoning? (2016)
2016
Cited alongside, same era.
Psychonomic Bulletin & Review
A. L. Cohen, S. Sidlowski, A. Staub, Beliefs and bayesian reasoning · 2017
Cited alongside, same era.
H. Mercier, D. Sperber, The Enigma of Reason
2017
Cited alongside, same era.
arXiv preprint arXiv:1709.08568
Original
Y. Bengio, The consciousness prior · 2017
Cited alongside, same era.
D. J. Benjamin, Errors in probabilistic reasoning and judgment biases, Tech. rep
2018
Cited alongside, same era.
arXiv preprint arXiv:1806.02847
Original
T. H. Trinh, Q. V. Le, A simple method for commonsense reasoning · 2018
Cited alongside, same era.
arXiv preprint arXiv:1802.04302
Original
I. Dasgupta, D. Guo, A. Stuhlmüller, S. J. Gershman, N. D. Goodman, Evaluating compositionality in sentence embeddings · 2018
Cited alongside, same era.

Then

A. Søgaard, Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
2021
Later among the works it cites.
Advances in Neural Information Processing Systems
M. Nye, M. Tessler, J. Tenenbaum, B. M. Lake, Improving coherence and consistency in neural sequence models with dual-system, neuro-symbolic reasoning · 2021
Later among the works it cites.
arXiv preprint arXiv:2104.01490
Original
R. Cao, D. Yamins, Explanatory models in neuroscience: Part 1–taking mechanistic abstraction seriously · 2021
Later among the works it cites.
arXiv preprint arXiv:2104.01489
Original
R. Cao, D. Yamins, Explanatory models in neuroscience: Part 2–constraint-based intelligibility · 2021
Later among the works it cites.
Proceedings of the National Academy of Sciences
M. Schrimpf, I. A. Blank, G. Tuckute, C. Kauf, E. A. Hosseini, N. Kanwisher, J. B. Tenenbaum, E. Fedorenko, The neural architecture of language: Integrative modeling converges on predictive processing · 2021
Later among the works it cites.
arXiv preprint arXiv:2109.01247
Original
A. Webson, E. Pavlick, Do prompt-based models really understand the meaning of their prompts? · 2021
Later among the works it cites.
arXiv preprint arXiv:2104.08773
Original
S. Mishra, D. Khashabi, C. Baral, H. Hajishirzi, Cross-task generalization via natural language crowdsourcing instructions · 2021
Later among the works it cites.
arXiv preprint arXiv:2107.06994
Original
A. J. Nam, J. L. McClelland, What underlies rapid learning and systematic generalization in humans · 2021
Later among the works it cites.
Y. Wu, M. N. Rabe, W. Li, J. Ba, R. B. Grosse, C. Szegedy, International Conference on Machine Learning
2021
Later among the works it cites.
Transactions of the Association for Computational Linguistics
T. Schick, S. Udupa, H. Schütze, Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP · 2021
Later among the works it cites.
arXiv preprint arXiv:2110.14168
Original
K. Cobbe, V. Kosaraju, M. Bavarian, J. Hilton, R. Nakano, C. Hesse, J. Schulman, Training verifiers to solve math word problems · 2021
Later among the works it cites.
Current Opinion in Behavioral Sciences
J. X. Wang, Meta-learning in natural and artificial intelligence · 2021
Later among the works it cites.
J. Dodge, M. Sap, A. Marasović, W. Agnew, G. Ilharco, D. Groeneveld, M. Mitchell, M. Gardner, Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
2021
Later among the works it cites.
arXiv preprint arXiv:2104.08315
Original
A. Holtzman, P. West, V. Shwartz, Y. Choi, L. Zettlemoyer, Surface form competition: Why the highest probability answer isn’t always right · 2021
Later among the works it cites.
A. Hosseini, S. Reddy, D. Bahdanau, R. D. Hjelm, A. Sordoni, A. Courville, Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
2021
Later among the works it cites.
arXiv preprint arXiv:2202.07785
Original
D. Ganguli, D. Hernandez, L. Lovitt, N. DasSarma, T. Henighan, A. Jones, N. Joseph, J. Kernion, B. Mann, A. Askell, et al · 2022
Closest in time.
arXiv preprint arXiv:2205.11916
Original
T. Kojima, S. S. Gu, M. Reid, Y. Matsuo, Y. Iwasawa, Large language models are zero-shot reasoners · 2022
Closest in time.
arXiv preprint arXiv:2206.07682
Original
J. Wei, Y. Tay, R. Bommasani, C. Raffel, B. Zoph, S. Borgeaud, D. Yogatama, M. Bosma, D. Zhou, D. Metzler, et al · 2022
Closest in time.
arXiv preprint arXiv:2202.07206
Original
Y. Razeghi, R. L. Logan IV, M. Gardner, S. Singh, Impact of pretraining term frequencies on few-shot reasoning · 2022
Closest in time.
K. Valmeekam, A. Olmo, S. Sreedharan, S. Kambhampati, Large language models still can’t plan (a benchmark for llms on planning and reasoning about change) (2022)
2022
Closest in time.
arXiv preprint arXiv:2203.15556
Original
J. Hoffmann, S. Borgeaud, A. Mensch, E. Buchatskaya, T. Cai, E. Rutherford, D. d. L. Casas, L. A. Hendricks, J. Welbl, A. Clark, et al · 2022
Closest in time.
Topics in Cognitive Science
M. H. Tessler, J. B. Tenenbaum, N. D. Goodman, Logic, probability, and pragmatics in syllogistic reasoning · 2022
Closest in time.
S. Kadavath, T. Conerly, A. Askell, T. Henighan, D. Drain, E. Perez, N. Schiefer, Z. H. Dodds, N. DasSarma, E. Tran-Johnson, S. Johnston, S. El-Showk, A. Jones, N. Elhage, T. Hume, A. Chen, Y. Bai, S. Bowman, S. Fort, D. Ganguli, D. Hernandez, J. Jacobson, J. Kernion, S. Kravec, L. Lovitt, K. Ndousse, C. Olsson, S. Ringer, D. Amodei, T. Brown, J. Clark, N. Joseph, B. Mann, S. McCandlish, C. Olah, J. Kaplan, Language models (mostly) know what they know (2022)
2022
Closest in time.
M. Binz, E. Schulz, Using cognitive psychology to understand gpt-3 (2022)
2022
Closest in time.
arXiv preprint arXiv:2201.11903
Original
J. Wei, X. Wang, D. Schuurmans, M. Bosma, E. Chi, Q. Le, D. Zhou, Chain of thought prompting elicits reasoning in large language models · 2022
Closest in time.
D. Khashabi, C. Baral, Y. Choi, H. Hajishirzi, Findings of the Association for Computational Linguistics: ACL 2022
2022
Closest in time.
arXiv preprint arXiv:2204.02329
Original
A. K. Lampinen, I. Dasgupta, S. C. Chan, K. Matthewson, M. H. Tessler, A. Creswell, J. L. McClelland, J. X. Wang, F. Hill, Can language models learn from explanations in context? · 2022
Closest in time.
arXiv preprint arXiv:2203.11171
Original
X. Wang, J. Wei, D. Schuurmans, Q. Le, E. Chi, D. Zhou, Self-consistency improves chain of thought reasoning in language models · 2022
Closest in time.
PLOS Computational Biology
Y. Li, J. L. McClelland, A weighted constraint satisfaction approach to human goal-directed decision making · 2022
Closest in time.
Nature neuroscience
A. Goldstein, Z. Zada, E. Buchnik, M. Schain, A. Price, B. Aubrey, S. A. Nastase, A. Feder, D. Emanuel, A. Cohen, et al · 2022
Closest in time.
arXiv preprint arXiv:2206.02885
Original
D. Schlangen, Norm participation grounds language · 2022
Closest in time.
arXiv preprint arXiv:2206.05802
Original
W. Saunders, C. Yeh, J. Wu, S. Bills, L. Ouyang, J. Ward, J. Leike, Self-critiquing models for assisting human evaluators · 2022
Closest in time.
arXiv preprint arXiv:2203.14465
Original
E. Zelikman, Y. Wu, N. D. Goodman, Star: Bootstrapping reasoning with reasoning · 2022
Closest in time.
Advances in Neural Information Processing Systems
S. Chan, A. Santoro, A. Lampinen, J. Wang, A. Singh, P. Richemond, J. McClelland, F. Hill, Data distributional properties drive emergent in-context learning in transformers · 2022
Closest in time.
Y. Tay, M. Dehghani, V. Q. Tran, X. Garcia, J. Wei, X. Wang, H. W. Chung, D. Bahri, T. Schuster, S. Zheng, et al
2022
Closest in time.
arXiv preprint arXiv:2304.15004
Original
R. Schaeffer, B. Miranda, S. Koyejo, Are emergent abilities of large language models a mirage? · 2023
Closest in time.
Gpt-3.5, https://platform.openai.com/docs/models/gpt-3-5
2023
Closest in time.
arXiv preprint arXiv:2305.10403
Original
R. Anil, A. M. Dai, O. Firat, M. Johnson, D. Lepikhin, A. Passos, S. Shakeri, E. Taropa, P. Bailey, Z. Chen, et al · 2023
Closest in time.
arXiv preprint arXiv:2304.03843
Original
B. Prystawski, N. D. Goodman, Why think step-by-step? reasoning emerges from the locality of experience · 2023
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…