Search

Your search keyword '"Schwitzgebel, David"' showing total 15 results
Start Over Author "Schwitzgebel, David"
15 results on '"Schwitzgebel, David"'

1. Creating a Large Language Model of a Philosopher

Author

Schwitzgebel, Eric, Schwitzgebel, David, and Strasser, Anna

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

Can large language models be trained to produce philosophical texts that are difficult to distinguish from texts produced by human philosophers? To address this question, we fine-tuned OpenAI's GPT-3 with the works of philosopher Daniel C. Dennett as additional training data. To explore the Dennett model, we asked the real Dennett ten philosophical questions and then posed the same questions to the language model, collecting four responses for each question without cherry-picking. We recruited 425 participants to distinguish Dennett's answer from the four machine-generated answers. Experts on Dennett's work (N = 25) succeeded 51% of the time, above the chance rate of 20% but short of our hypothesized rate of 80% correct. For two of the ten questions, the language model produced at least one answer that experts selected more frequently than Dennett's own answer. Philosophy blog readers (N = 302) performed similarly to the experts, while ordinary research participants (N = 98) were near chance distinguishing GPT-3's responses from those of an "actual human philosopher"., Comment: Earlier version (Nov 2, 2022) available at http://www.faculty.ucr.edu/~eschwitz/SchwitzPapers/GPT-3-Dennett-221102.pdf; stable URL for updated versions http://faculty.ucr.edu/~eschwitz/SchwitzAbs/GPT3Dennett.htm

Published
2023

2. Accessibility of Agent and Patient roles is shaped by context

Author

Schwitzgebel, David, Rissman, Lilia, and Strickland, Brent

Subjects
Concepts and categories, Event cognition, Language and thought, Learning, Semantics of language
Abstract

Adults represent events through abstract roles: e.g., when Jim eats chocolate, Jim is an “Agent” and the chocolate a “Patient.” These roles are thought to be universally prominent across languages (Strickland, 2016). Nonetheless, learning to sort pictures into Agent and Patient categories is difficult for some individuals (Rissman & Lupyan, 2022). Which contexts affect the accessibility of these roles to explicit reasoning? Across two pre-registered experiments, English-speaking adults viewed pictures of an Agent acting on a Patient (e.g., one person kicking another). In each picture, a red dot marked either the Agent or the Patient. In one condition, participants freely sorted the pictures into two categories. In another condition, the pictures were accompanied by a nonsense verb marked with the suffix -ja or -ka. In both experiments, participants in the language condition were significantly more likely to sort by the Agent/Patient distinction, suggesting that language makes abstract roles more accessible.

Published
2023

3. Creating a large language model of a philosopher.

Author

Schwitzgebel, Eric, Schwitzgebel, David, and Strasser, Anna

Subjects
*LANGUAGE models, *GENERATIVE pre-trained transformers, *PHILOSOPHERS, *ARTIFICIAL intelligence
Abstract

Can large language models produce expert‐quality philosophical texts? To investigate this, we fine‐tuned GPT‐3 with the works of philosopher Daniel Dennett. To evaluate the model, we asked the real Dennett 10 philosophical questions and then posed the same questions to the language model, collecting four responses for each question without cherry‐picking. Experts on Dennett's work succeeded at distinguishing the Dennett‐generated and machine‐generated answers above chance but substantially short of our expectations. Philosophy blog readers performed similarly to the experts, while ordinary research participants were near chance distinguishing GPT‐3's responses from those of an "actual human philosopher". [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

7. Rapid judgment of social scenes in a linguistic priming task, masks version

Author

Schwitzgebel, David, Falck, Andreas, Strickland, Brent, Papeo, Liuba, and Wittenberg, Eva

Subjects
FOS: Psychology, Cognition and Perception, Engineering, FOS: Languages and literature, Vision Science, Psychology, Linguistics, priming, inversion effects, events, language, visual perception, Social and Behavioral Sciences, Biomedical Engineering and Bioengineering
Abstract

This is a pre-registration for the third study within the broader project. This experiment aims to determine whether the prime-related effects observed in Experiment 1 are confounded by task structure. This experiment utilizes randomly-generated masks in place of the word primes.

Published
2022
Full Text
View/download PDF

8. Rapid judgment of social scenes in a linguistic priming task, single-bodies version

Author

Schwitzgebel, David, Falck, Andreas, Strickland, Brent, Papeo, Liuba, and Wittenberg, Eva

Subjects
FOS: Psychology, Cognition and Perception, Engineering, FOS: Languages and literature, Vision Science, Psychology, Linguistics, priming, inversion effects, events, language, visual perception, Social and Behavioral Sciences, Biomedical Engineering and Bioengineering
Abstract

This is a pre-registration for the second study within the broader project. This experiment aims to determine whether the prime-related effects observed in Experiment 1 generalize to single bodies. This experiment utilizes images with single bodies/chairs in place of images with dyads of bodies/chairs.

Published
2022
Full Text
View/download PDF

12. PONG: A Task-Demand Study

Author

Sample, Lindsey, Carreras, Joshua, Khalsa, Nanak, Schwitzgebel, David, and Rich, Eliana

Abstract

Behavioral Study by Eliana, David, Nanak, Josh and Lindsey

Published
2022
Full Text
View/download PDF

13. An Event-Related Potential Study of Emotion-Induced Blindness

Author

de Leeuw, Joshua, Livingston, Ken, Alexandrov, Emma, Andrews, Rebecca, Beech, Caroline, Berger, Noe, Carreras, Joshua, Catling-Allen, Daisy, Chen, Jennifer, Geng, Ningyao, Grossman, Katya, Guo, Wenxuan, Holm, Kiara, Landolt, Cole, Li, Jessica, Marks, Parker, Khalsa, Nanak, Morris, Noelle, Pan, Dingyi, Rich, Eliana, Sample, Lindsey, Schmitt, Maddy, Shihan, Zhao, and Schwitzgebel, David

Subjects
genetic structures
Abstract

This study is designed to replicate the Kennedy et al. (2014) study that tests whether emotion-induced blindness has a similar neural signature to attentional blink (AB). Event-related potentials (specifically N2 and P3b) will be used to investigate the effect of emotional distractors on early and late processing stages.

Published
2022
Full Text
View/download PDF

14. Core Categories in Language: Interactions between Semantic Categories and Syntactic Positioning

Author

Schwitzgebel, David

Subjects
Social and Behavioral Sciences
Abstract

Previous studies from our research group (O’Madagain, Bai, Peperkamp, & Strickland, 2019) identified a novel connection between language learning and semantic categories: certain anthropocentric distinctions (e.g., animacy, biological sex, humanness), when encoded in a grammatical affix, are easier to learn in an artificial language than non-anthropocentric distinctions (e.g., “vehicleness”). This applies even when the distinctions are at the same level of categorical abstraction (superordinate, basic, or subordinate). Therefore, these distinctions seem to operate as a “core” part of language evolution. However, these studies only investigated affixes positioned in the noun phrase of the artificial language. In this study, we aim to extend the previous methodology to artificial languages in which affixes are positioned in the verb phrase. This will enable us to investigate the possible relationship between syntactic positioning, “core” semantic categories, and language learning.

Published
2022
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results