How cognitive selection affects language change.

Like biological species, words in language must compete to survive. Previously, it has been shown that language changes in response to cognitive constraints and over time becomes more learnable. Here, we use two complementary research paradigms to demonstrate how the survival of existing word forms can be predicted by psycholinguistic properties that impact language production. In the first study, we analyzed the survival of words in the context of interpersonal communication. We analyzed data from a large-scale serial-reproduction experiment in which stories were passed down along a transmission chain over multiple participants. The results show that words that are acquired earlier in life, more concrete, more arousing, and more emotional are more likely to survive retellings. We reason that the same trend might scale up to language evolution over multiple generations of natural language users. If that is the case, the same set of psycholinguistic properties should also account for the change of word frequency in natural language corpora over historical time. That is what we found in two large historical-language corpora (Study 2): Early acquisition, concreteness, and high arousal all predict increasing word frequency over the past 200 y. However, the two studies diverge with respect to the impact of word valence and word length, which we take up in the discussion. By bridging micro-level behavioral preferences and macro-level language patterns, our investigation sheds light on the cognitive mechanisms underlying word competition.

Historical language records reveal a surge of cognitive distortions in recent decades.

Individuals with depression are prone to maladaptive patterns of thinking, known as cognitive distortions, whereby they think about themselves, the world, and the future in overly negative and inaccurate ways. These distortions are associated with marked changes in an individual's mood, behavior, and language. We hypothesize that societies can undergo similar changes in their collective psychology that are reflected in historical records of language use. Here, we investigate the prevalence of textual markers of cognitive distortions in over 14 million books for the past 125 y and observe a surge of their prevalence since the 1980s, to levels exceeding those of the Great Depression and both World Wars. This pattern does not seem to be driven by changes in word meaning, publishing and writing standards, or the Google Books sample. Our results suggest a recent societal shift toward language associated with cognitive distortions and internalizing disorders.

High-stakes decisions do not require narrative conviction but narrative flexibility.

We challenge Johnson et al.'s assumption that people reduce unclear situations to a single narrative explanation and that such reduction would be adaptive for decision-making under radical uncertainty. Instead, we argue that people imagine and maintain multiple narrative possibilities throughout the decision-making process and that this process provides cognitive flexibility and adaptive benefits within the proposed model.

Serial reproduction of narratives preserves emotional appraisals.

We conducted the largest multiple-iteration retelling study to date (12,840 participants and 19,086 retellings) with two different studies that test how emotional appraisals are transmitted across retellings. We use a novel Bayesian model that tracks changes across retellings. Study 1 examines the preservation of appraisals of happy and sad stories and finds that retellings preserve the story's degree of happiness and sadness even when length shrinks and aspects of story coherence and rationalisation deteriorate. Study 2 compared the transmission of appraisals of happiness and sadness with embarrassment, disgust, and risk. Appraisals of happiness, sadness, and also embarrassment showed high appraisal preservation, while disgust and risk were not well preserved. We conclude that participants in our studies encoded happy and sad stories by encapsulating the events and details into an overall emotional appraisal of the story and that this processing strategy might also apply to stories involving other emotions like embarrassment. The emotional appraisal played a key role in retelling by helping to guide the selection, invention, and ordering of the story elements. Hence, we posit that emotion appraisals can operate as anchors for remembering and retelling stories, thus playing an important role in narrative communication.

Humans create more novelty than ChatGPT when asked to retell a story.

We compare how humans retell stories to how ChatGPT retells stories in chains of three retellings by different people or different accounts on ChatGPT. ChatGPT provides competent summaries of the original narrative texts in one step of retelling. In subsequent retellings few additional changes occur. Human retellers, by contrast, reduce the original text incrementally and by creating 55-60% of novel words and concepts (synsets) at each iteration. The retellings by both ChatGPT and humans show very stable emotion ratings, which is a puzzle for human retellers given the high degree of novel inventions across retellings. ChatGPT maintains more nouns, adjectives, and prepositions and also uses language later acquired in life, while humans use more verbs, adverbs, and negations and use language acquired at a younger age. The results reveal that spontaneous retelling by humans involves ongoing creativity, anchored by emotions, beyond the default probabilistic wording of large language models such as ChatGPT.

Quantifying the retention of emotions across story retellings.

Story retelling is a fundamental medium for the transmission of information between individuals and among social groups. Besides conveying factual information, stories also contain affective information. Though natural language processing techniques have advanced considerably in recent years, the extent to which machines can be trained to identify and track emotions across retellings is unknown. This study leverages the powerful RoBERTa model, based on a transformer architecture, to derive emotion-rich story embeddings from a unique dataset of 25,728 story retellings. The initial stories were centered around five emotional events (joy, sadness, embarrassment, risk, and disgust-though the stories did not contain these emotion words) and three intensities (high, medium, and low). Our results indicate (1) that RoBERTa can identify emotions in stories it was not trained on, (2) that the five emotions and their intensities are preserved when they are transmitted in the form of retellings, (3) that the emotions in stories are increasingly well-preserved as they experience additional retellings, and (4) that among the five emotions, risk and disgust are least well-preserved, compared with joy, sadness, and embarrassment. This work is a first step toward quantifying situation-driven emotions with machines.

Choosing and enjoying violence in narratives.

We use an interactive story design in which participants read short stories and make two consecutive plot choices about whether protagonists commit low- or high-violence actions. Our study has four main findings. 1) People who choose high violence report greater satisfaction with the story, while those switching to or staying with no violence show lower satisfaction. 2) However, when participants encounter these stories without choices, they reliably rate higher-violence stories as less satisfying than lower-violence stories. 3) Regret seems to account for the low satisfaction of those who choose or switch to low violence. 4) There is a large segment of people (up to 66%) who can be persuaded by different story contexts (genre, perspective) to choose extreme violence in interactive fiction and as a consequence of their choice feel satisfaction. We hypothesize that people who opt for high violence enjoy the story as a result of their choice. Overall, we suggest that choosing violence serves as a gateway for enjoyment by creating an aesthetic zone of control detached from morality.

Fact vs. Affect in the Telephone Game: All Levels of Surprise Are Retold With High Accuracy, Even Independently of Facts.

When people retell stories, what guides their retelling? Most previous research on story retelling and story comprehension has focused on information accuracy as the key measure of stability in transmission. This paper suggests that there is a second, affective, dimension that provides stability for retellings, namely the audience affect of surprise. In a large-sample study with multiple iterations of retellings, we found evidence that people are quite accurate in preserving all degrees of surprisingness in serial reproduction - even when the event that produced the surprisingness in the original story is dropped or changed. Thus, we propose that the preservation of affect is an implicit goal of retelling: merely do retellers not recall highly surprising events better, but rather they register all levels of surprisingness precisely and aim to surprise their implied audience to same degree. This study used 2,389 participants. Significance Statement: Story retelling is a process whereby cultural information is transmitted horizontally across social networks and vertically down generations. For the most part, retelling research has focused on the relevance and stability of factual information, "who did what, where, when, and why"; comparatively little is known about the transmission of affective information. We suggest that affect can serve as a second axis of stability for retelling, partially independent from factual information. In serial reproduction tasks modeled after the telephone game, we find that surprisingness of stories is well preserved across retellings - even when the facts and events of the story are not. The findings are significant for the communication of information, and thereby also the stability and transformation of culture in general.