Abstract

Cryptocurrency markets have reached an unprecedented scale and mainstream acceptance, with Bitcoin, boasting a market cap of $2.24 trillion, hitting new all-time highs above $124,290 in August 2025, while institutional integration continues to accelerate globally. Despite this growth, Bitcoin's volatility remains at 18.37% as of mid-2025. Consequently, cryptocurrency markets present a decision environment characterized by extreme volatility, fragmented information, and significant emotional impact. However, decision support systems (DSS) used by investors and traders primarily focus on rational-analytic data – offering financial products through cost-effective technical methods – while overlooking the affective drivers of behavior. In other words, existing financial DSS emphasize cognitive-rational analysis and neglect affective inputs. Within the Information Systems and Financial Technology literature, this creates a significant gap in current DSS design and behavioral decision-making. The predominance of cognitive-rational DSS is especially problematic in volatile environments where fast, intuitive, affect-driven sentiments often dominate. Drawing on Dual-Process Theory (DPT) and Information Processing Theory (IPT), we propose a sentiment-augmented DSS that integrates affective-cognitive information processing. We argue that explicit sentiment cues embedded in DSS can prompt decision-makers to shift from fast, intuitive (System 1) to deliberative (System 2) processing, thereby enhancing the quality of their decisions in a highly volatile environment. Dual-Process Theory distinguishes between System 1 (fast, intuitive, affect-driven) and System 2 (slow, deliberative, effortful) cognition. In high-volatility contexts such as cryptocurrency markets, System 1 often dominates, leading to suboptimal decisions. Embedding sentiment cues can act as decision nudges, encouraging System 2 engagement when market sentiment is extreme. Furthermore, information processing theory posits that decision quality depends on the availability of bounded cognitive resources and the structured presentation of information. Sentiment cues – whether positive or negative – can provide contextual framing or the needed bounded cognitive resource, reducing information search complexity and reallocating attention efficiently, thereby improving decision performance. We will test this proposition through a laboratory experiment (n = 124) with outcome measures (decision quality and portfolio performance), psychophysiological measurements (EEG, eye-tracking), and process measures (cognitive load, information processing depth, and behavioral bias measurement). We employ a 3×3 mixed-factorial experimental design, with sentiment display condition as a between-subjects factor (control, numeric sentiment, i.e., based on the Fear & Greed Index, and contextual sentiment with textual interpretation), and market regime as a within-subjects factor (Bull, Bear, Sideways). Our conceptual model predicts that sentiment cues will reduce cognitive load, enhance information processing, and improve decision quality. We expect to find that users of sentiment-augmented DSS demonstrate higher decision quality and portfolio returns than control DSS users. Mediation analyses are expected to reveal that reduced cognitive load and deeper information processing explain these effects. Additionally, we anticipate that further analysis will reveal that sentiment-augmented DSS can mitigate biases such as the disposition effect, herding, and overconfidence. We summarize the paper’s expected contribution as follows: (1) an extension of Dual-Process Theory, showing how system design can prompt deliberative processing in financial decision-making. (2) an advancement of Information Processing Theory by illustrating affective cues as load-reducing, framing tools. (3) the introduction of a model of affective-cognitive DSS design suited for high-emotion decision environments beyond finance (e.g., healthcare, crisis management).

Download

Abstract Only

Share

COinS