# Mechanisms for Handling Nested Dependencies in Neural-Network Language Models and Humans

arXiv: https://arxiv.org/abs/2006.11098 · HTML: https://mlsanity.com/paper/2006.11098/

## Key facts

- Cited by 42 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Jun 2020.

## Built on by (most cited first, 30 of 42)

- [Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models](https://mlsanity.com/paper/2301.04213/) (2023)
- [Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little](https://mlsanity.com/paper/2104.06644/) (2021)
- [Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models](https://mlsanity.com/paper/2106.06087/) (2021)
- [State-of-the-art generalisation research in NLP: A taxonomy and review](https://mlsanity.com/paper/2210.03050/) (2022)
- [Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Prediction](https://mlsanity.com/paper/2105.06965/) (2021)
- [Structural Persistence in Language Models: Priming as a Window into Abstract Language Representations](https://mlsanity.com/paper/2109.14989/) (2021)
- [On the proper role of linguistically-oriented deep net analysis in linguistic theorizing](https://mlsanity.com/paper/2106.08694/) (2021)
- [Can language models handle recursively nested grammatical structures? A case study on comparing models and humans](https://mlsanity.com/paper/2210.15303/) (2022)
- [Language acquisition: do children and language models follow similar learning stages?](https://mlsanity.com/paper/2306.03586/) (2023)
- [Sparse Interventions in Language Models with Differentiable Masking](https://mlsanity.com/paper/2112.06837/) (2021)
- [The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling](https://mlsanity.com/paper/2410.09223/) (2024)
- [A polar coordinate system represents syntax in large language models](https://mlsanity.com/paper/2412.05571/) (2024)
- [How to Plant Trees in Language Models: Data and Architectural Effects on the Emergence of Syntactic Inductive Biases](https://mlsanity.com/paper/2305.19905/) (2023)
- [Learned feature representations are biased by complexity, learning order, position, and more](https://mlsanity.com/paper/2405.05847/) (2024)
- [A Language Model with Limited Memory Capacity Captures Interference in Human Sentence Processing](https://mlsanity.com/paper/2310.16142/) (2023)
- [Language in Vivo vs. in Silico: Size Matters but Larger Language Models Still Do Not Comprehend Language on a Par with Humans Due to Impenetrable Semantic Reference](https://mlsanity.com/paper/2404.14883/) (2024)
- [Kallini et al. (2024) do not compare impossible languages with constituency-based ones](https://mlsanity.com/paper/2410.12271/) (2024)
- [Causal Analysis of Syntactic Agreement Neurons in Multilingual Language Models](https://mlsanity.com/paper/2210.14328/) (2022)
- [A Psycholinguistic Evaluation of Language Models' Sensitivity to Argument Roles](https://mlsanity.com/paper/2410.16139/) (2024)
- [Information-Restricted Neural Language Models Reveal Different Brain Regions' Sensitivity to Semantics, Syntax and Context](https://mlsanity.com/paper/2302.14389/) (2023)
- [Blackbird language matrices (BLM), a new task for rule-like generalization in neural networks: Motivations and Formal Specifications](https://mlsanity.com/paper/2306.11444/) (2023)
- [A systematic framework for generating novel experimental hypotheses from language models](https://mlsanity.com/paper/2408.05086/) (2024)
- [Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans](https://mlsanity.com/paper/2110.07240/) (2021)
- [Characterizing Verbatim Short-Term Memory in Neural Language Models](https://mlsanity.com/paper/2210.13569/) (2022)
- [Different types of syntactic agreement recruit the same units within large language models](https://mlsanity.com/paper/2512.03676/) (2025)
- [Assessing the Capacity of Transformer to Abstract Syntactic Representations: A Contrastive Analysis Based on Long-distance Agreement](https://mlsanity.com/paper/2212.04523/) (2022)
- [Causal interventions expose implicit situation models for commonsense language understanding](https://mlsanity.com/paper/2306.03882/) (2023)
- [From Form(s) to Meaning: Probing the Semantic Depths of Language Models Using Multisense Consistency](https://mlsanity.com/paper/2404.12145/) (2024)
- [Do LSTMs See Gender? Probing the Ability of LSTMs to Learn Abstract Syntactic Rules](https://mlsanity.com/paper/2211.00153/) (2022)
- [Large language models and (non-)linguistic recursion](https://mlsanity.com/paper/2306.07195/) (2023)

## Abstract (arXiv)

Recursive processing in sentence comprehension is considered a hallmark of human linguistic abilities. However, its underlying neural mechanisms remain largely unknown. We studied whether a modern artificial neural network trained with "deep learning" methods mimics a central aspect of human sentence processing, namely the storing of grammatical number and gender information in working memory and its use in long-distance agreement (e.g., capturing the correct number agreement between subject and verb when they are separated by other phrases). Although the network, a recurrent architecture with Long Short-Term Memory units, was solely trained to predict the next word in a large corpus, analysis showed the emergence of a very sparse set of specialized units that successfully handled local and long-distance syntactic agreement for grammatical number. However, the simulations also showed that this mechanism does not support full recursion and fails with some long-range embedded dependencies. We tested the model's predictions in a behavioral experiment where humans detected violations in number agreement in sentences with systematic variations in the singular/plural status of multiple nouns, with or without embedding. Human and model error patterns were remarkably similar, showing that the model echoes various effects observed in human data. However, a key difference was that, with embedded long-range dependencies, humans remained above chance level, while the model's systematic errors brought it below chance. Overall, our study shows that exploring the ways in which modern artificial neural networks process sentences leads to precise and testable hypotheses about human linguistic performance.