Why do you ask? The informational dynamics of questions and answers

Abstract

Asking questions is one of our most efficient and reliable means of learning about the world. Yet we do not often pose these questions to an impartial oracle; we ask cooperative social partners, in dialogue. In this paper, we aim to reconcile formal models of optimal question asking and answering with classic effects of social context. We begin from the observation that question-answer dialogue is motivated by a two-sided asymmetry in beliefs: questioners have a private goal but lack goal-relevant information about the world, and answerers have private information but lack knowledge about the questioner's goal. We formalize this problem in a computational framework and derive pragmatic questioner and answerer behavior from recursive social reasoning. Critically, we predict that pragmatic answerers go beyond the literal meaning of the question to be informative with respect to inferred goals, and that pragmatic questioners may therefore select questions to more unambiguously signal their goals. We evaluate our pragmatic models against asocial models in two ways. First, we present computational simulations accounting for three classic answerer effects in psycholinguistics. We then introduce the Hidden Goal paradigm for experimentally eliciting questioner and answerer behavior in scenarios where there is uncertainty about the questioner's goal. We report data from three experiments in this paradigm and show how our core computational framework can be composed with more sophisticated question semantics, hierarchical goal spaces, and a persistent state over which extended dialogue can unfold. We find that social inference is needed to account for critical aspects of the data.

Color Overmodification Emerges from Data-Driven Learning and Pragmatic Reasoning

“…The Rational Speech Acts (RSA) framework is a family of computational models of communication (Frank & Goodman, 2012) where two rational agents, a speaker and a listener, are modeled as recursively reasoning about each other when producing and interpreting utterances. This work follows important precedents from game theory (Lewis, 1969;Jager, 2007;Franke, 2009;Golland et al, 2010), and has been shown to capture a wide range of linguistic phenomena (e.g., Goodman and Stuhlmüller, 2013;Kao et al, 2014;Hawkins and Goodman, 2017;van Tiel et al, 2021).…”

Reconciling truthfulness and relevance as epistemic and decision-theoretic utility.

“…It effectively coarse-grains the set of possible worlds, allowing the speaker and listener to ignore irrelevant details. RSA-based models have integrated the notion of a QUD (Goodman & Lassiter, 2015) to explain interpretation of nonliteral or vague language in various forms (Hawkins et al, 2015;Kao & Goodman, 2015;Kao, Wu, et al, 2014;Lassiter & Goodman, 2017;Yoon et al, 2020). This document is copyrighted by the American Psychological Association or one of its allied publishers.…”

“…Conversely, speakers can use the same principles to select information that maximally improves the listener's decision making (Benz, 2006;Benz & Van Rooij, 2007). This can be seen when speakers "go beyond" the literal content of a question and supply additional decision-relevant knowledge (Hawkins et al, 2015). For example, asked if they accepted credit cards, restaurant owners answered the question and then spontaneously informed the caller about unexpected closures: "Uh, yes, we accept credit cards.…”

Pragmatic Issue-Sensitive Image Captioning

“…Our key modeling innovation lies in building issues into these models. In this, we are inspired by linguistic work on question-sensitive RSA (Goodman and Lassiter, 2015;Hawkins and Goodman, 2019).…”

“…Each cell represents a possible resolution of the issue. These ideas are brought into RSA by Goodman and Lassiter (2015) and Hawkins and Goodman (2019). We translate those ideas into the models for ISIC (Section 4), where an issue takes the form of a partition over a set of natural images.…”