Jakub Szymanik
Biographic Data
| ID | 5911020 |
|---|---|
| NAME | Jakub Szymanik |
| GIVEN NAMES | Jakub |
| FAMILY NAME | Szymanik |
| SIGNATURE | SZYMANIK J |
| AFFILIATIONS | University of Amsterdam |
| ORCID | 0000-0002-6145-6322 |
| VERIFIED | Yes |
| TOTAL WORKS | 16 |
| TOTAL CITATIONS | 1 |
| WORKS AS AUTHOR | 16 |
| WORKS AS EDITOR | 0 |
| FIRST PUBLICATION YEAR | 2009 |
| LATEST PUBLICATION YEAR | 2026 |
| H-INDEX | 1 |
Adult learning of a novel quantifier tracks semantic universals
Natural languages share common properties called universals. In the domain of quantification, three semantic universals were discovered: monotonicity (convexity), quantity, and conservativity. Researchers have been trying to explain the origin of semantic universals for decades. In this study, we tested one of the proposed explanations, the learnability hypothesis. According to this hypothesis, quantifiers that satisfy universals are easier to le…
Most quantifiers have many meanings
In this paper, we investigate, by means of a computational model, how individuals map quantifiers onto numbers and how they order quantifiers on a mental line. We selected five English quantifiers ( few , fewer than half , many , more than half , and most ) which differ in truth conditions and vagueness. We collected binary truth value judgment data in an online quantifier verification experiment. Using a Bayesian three-parameter logistic regress…
Quantifiers satisfying semantic universals have shorter minimal description length
Despite wide variation among natural languages, there are linguistic properties thought to be universal to all or nearly all languages. Here, we consider universals at the semantic level, in the domain of quantifiers, which are given by the properties of monotonicity, quantity, and conservativity, and we investigate whether these universals might be explained by differences in complexity. First, we use a minimal pair methodology and compare the c…
The Boolean Language of Thought is recoverable from learning data
Are Most and More Than Half Truth-Conditionally Equivalent?
Quantifying determiners most and more than half are standardly assumed to have the same truth-conditional meaning. Much work builds on this assumption in studying how the two quantifiers are mentally encoded and processed (Hackl, 2009; Lidz et al., 2011; Pietroski et al., 2009; Steinert-Threlkeld et al., 2015; Szymanik & Zajenkowski, 2010; Talmina et al., 2017). There is however empirical evidence that most is sometimes interpreted as ‘significan…
Neural Networks Track the Logical Complexity of Boolean Concepts
The language of thought hypothesis and connectionism provide two main accounts of category acquisition in the cognitive sciences. However, it is unclear to what extent their predictions agree. In this article, we tackle this problem by comparing the two accounts with respect to a common set of predictions about the effort required to acquire categories. We find that the two accounts produce similar predictions in the domain of Boolean categorizat…
Computational complexity explains neural differences in quantifier verification
Different classes of quantifiers provably require different verification algorithms with different complexity profiles. The algorithm for proportional quantifiers, like 'most', is more complex than that for nonproportional quantifiers, like 'all' and 'three'. We tested the hypothesis that different complexity profiles affect ERP responses during sentence verification, but not during sentence comprehension. In experiment 1, participants had to det…
An alternatives account of 'most’ and 'more than half’
While ‘most’ and ‘more than half’ are generally assumed to be truth-conditionally equivalent, the former is usually interpreted as conveying greater proportions than the latter. Previous work has attempted to explain this difference in terms of pragmatic strengthening or variation in meanings. In this paper, we propose a novel explanation that keeps the truth-conditions equivalence. We argue that the difference in typical sets between the two exp…
Conjoining Meanings
What is the relation between semantics and cognition? There are at least two natural approaches to this question. One may argue that the relationship between meaning and psychology is permissive, that is, meaning is externally specified and then in some way exported to our heads. On that view, it is often hard to systematically explain the connection between meaning and speakers’ judgments of entailment or the truth of sentences in context. Anoth…
Ease of learning explains semantic universals
Learnability and semantic universals
One of the great successes of the application of generalized quantifiers to natural language has been the ability to formulate robust semantic universals. When such a universal is attested, the question arises as to the source of the universal. In this paper, we explore the hypothesis that many semantic universals arise because expressions satisfying the universal are easier to learn than those that do not. While the idea that learnability explai…
Exploring the relation between semantic complexity and quantifier distribution in large corpora
Working Memory Mechanism in Proportional Quantifier Verification
Exploring the tractability border in epistemic tasks
Semantic bounds for everyday language
We consider the notion of everyday language. We claim that everyday language is semantically bounded by the properties expressible in the existential fragment of second-order logic. Two arguments for this thesis are formulated. First, we show that Barwise's so-called test of negation normality works properly only when assuming our main thesis. Second, we discuss the argument from practical computability for finite universes. Everyday language sen…
Branching Quantification v. Two-way Quantification
We discuss the thesis formulated by Hintikka (1973) that certain natural language sentences require non-linear quantification to express their meaning. We investigate sentences with combinations of quantifiers similar to Hintikka’s examples and propose a novel alternative reading expressible by linear formulae. This interpretation is based on linguistic and logical observations. We report on our experiments showing that people tend to interpret s…
Branching Quantification v. Two-way Quantification
We discuss the thesis formulated by Hintikka (1973) that certain natural language sentences require non-linear quantification to express their meaning. We investigate sentences with combinations of quantifiers similar to Hintikka’s examples and propose a novel alternative reading expressible by linear formulae. This interpretation is based on linguistic and logical observations. We report on our experiments showing that people tend to interpret s…
Semantic bounds for everyday language
We consider the notion of everyday language. We claim that everyday language is semantically bounded by the properties expressible in the existential fragment of second-order logic. Two arguments for this thesis are formulated. First, we show that Barwise's so-called test of negation normality works properly only when assuming our main thesis. Second, we discuss the argument from practical computability for finite universes. Everyday language sen…
Working Memory Mechanism in Proportional Quantifier Verification
Exploring the tractability border in epistemic tasks
Exploring the relation between semantic complexity and quantifier distribution in large corpora
Learnability and semantic universals
One of the great successes of the application of generalized quantifiers to natural language has been the ability to formulate robust semantic universals. When such a universal is attested, the question arises as to the source of the universal. In this paper, we explore the hypothesis that many semantic universals arise because expressions satisfying the universal are easier to learn than those that do not. While the idea that learnability explai…
Ease of learning explains semantic universals
An alternatives account of 'most’ and 'more than half’
While ‘most’ and ‘more than half’ are generally assumed to be truth-conditionally equivalent, the former is usually interpreted as conveying greater proportions than the latter. Previous work has attempted to explain this difference in terms of pragmatic strengthening or variation in meanings. In this paper, we propose a novel explanation that keeps the truth-conditions equivalence. We argue that the difference in typical sets between the two exp…
Conjoining Meanings
What is the relation between semantics and cognition? There are at least two natural approaches to this question. One may argue that the relationship between meaning and psychology is permissive, that is, meaning is externally specified and then in some way exported to our heads. On that view, it is often hard to systematically explain the connection between meaning and speakers’ judgments of entailment or the truth of sentences in context. Anoth…
Are Most and More Than Half Truth-Conditionally Equivalent?
Quantifying determiners most and more than half are standardly assumed to have the same truth-conditional meaning. Much work builds on this assumption in studying how the two quantifiers are mentally encoded and processed (Hackl, 2009; Lidz et al., 2011; Pietroski et al., 2009; Steinert-Threlkeld et al., 2015; Szymanik & Zajenkowski, 2010; Talmina et al., 2017). There is however empirical evidence that most is sometimes interpreted as ‘significan…
Neural Networks Track the Logical Complexity of Boolean Concepts
The language of thought hypothesis and connectionism provide two main accounts of category acquisition in the cognitive sciences. However, it is unclear to what extent their predictions agree. In this article, we tackle this problem by comparing the two accounts with respect to a common set of predictions about the effort required to acquire categories. We find that the two accounts produce similar predictions in the domain of Boolean categorizat…
Computational complexity explains neural differences in quantifier verification
Different classes of quantifiers provably require different verification algorithms with different complexity profiles. The algorithm for proportional quantifiers, like 'most', is more complex than that for nonproportional quantifiers, like 'all' and 'three'. We tested the hypothesis that different complexity profiles affect ERP responses during sentence verification, but not during sentence comprehension. In experiment 1, participants had to det…
Quantifiers satisfying semantic universals have shorter minimal description length
Despite wide variation among natural languages, there are linguistic properties thought to be universal to all or nearly all languages. Here, we consider universals at the semantic level, in the domain of quantifiers, which are given by the properties of monotonicity, quantity, and conservativity, and we investigate whether these universals might be explained by differences in complexity. First, we use a minimal pair methodology and compare the c…
The Boolean Language of Thought is recoverable from learning data
Most quantifiers have many meanings
In this paper, we investigate, by means of a computational model, how individuals map quantifiers onto numbers and how they order quantifiers on a mental line. We selected five English quantifiers ( few , fewer than half , many , more than half , and most ) which differ in truth conditions and vagueness. We collected binary truth value judgment data in an online quantifier verification experiment. Using a Bayesian three-parameter logistic regress…
Adult learning of a novel quantifier tracks semantic universals
Natural languages share common properties called universals. In the domain of quantification, three semantic universals were discovered: monotonicity (convexity), quantity, and conservativity. Researchers have been trying to explain the origin of semantic universals for decades. In this study, we tested one of the proposed explanations, the learnability hypothesis. According to this hypothesis, quantifiers that satisfy universals are easier to le…
Computer Science (14 works) · Artificial Intelligence (9 works) · Psychology (9 works) · Linguistics (7 works) · Mathematics (7 works) · Natural language processing (7 works) · Philosophy (7 works) · Epistemology (6 works) · Artificial Intelligence (5 works) · Child and Animal Learning Development (4 works)