Obsah: :: Library Catalog

Uloženo v:

Podrobná bibliografie
Hlavní autor:	Boullineau, Emile
Médium:	Recurso digital
Jazyk:
Vydáno:	Zenodo 2026
Témata:	emotional intelligence EQ four-branch model MSCEIT emotion perception misattribution rejection sensitivity hostile attribution bias empathic accuracy Theory of mind alexithymia appraisal theory constructionist emotion theory artificial intelligence AI AI emotion classification
On-line přístup:	https://doi.org/10.5281/zenodo.19962501
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Obsah:

This preprint introduces the Branch Dissociation Hypothesis: a refinement of the four-branch model of emotional intelligence that identifies a previously undertheorised profile within the EQ measurement tradition. The framework argues that detecting affective signals and attributing their causes are separable capacities, and that their dissociation produces a specific failure mode the paper calls interpretive miscalibration. The diagnostic profile is people who detect affective signals well but attribute their causes poorly, with high confidence in the wrong attribution. When the situation is ambiguous and the perceiver could plausibly be its cause, this profile may make emotional perceptiveness a liability rather than an asset, because confident misreading prompts action while non-detection prompts none. The paper proposes two reusable methodological tools. Branch-Residualised Interpretation is a scoring approach that isolates attribution accuracy from detection accuracy. The Self-Referential Attribution Task (SRAT) is a procedure designed to surface the diagnostic profile under self-relevant ambiguity, conditions standard EQ tasks have systematically stripped away. Eight numbered falsifiable predictions and an eight-item research agenda are offered. The framework is anchored in three convergent literatures (rejection sensitivity, hostile attribution bias, and paranoid cognition in non-clinical populations) and engaged with adjacent constructs (empathic accuracy, theory of mind, alexithymia, metacognitive monitoring, predictive processing, and appraisal theory). It is held compatible with the constructionist programme in emotion science. Practical implications span clinical practice, organisational and leadership selection, executive coaching, educational psychology, and assessment-industry reform. The paper also flags artificial systems performing emotion classification as a useful contrast case. Current AI systems and vision-language models lack the first-person self-implication mechanism that drives interpretive miscalibration in humans, though they may inherit other attributional biases from training data. The framework does not claim AI is generically more emotionally intelligent than humans; the human–AI gap on inferential affect-attribution under self-relevant ambiguity is itself diagnostic of where in the EQ architecture self-referential bias bites. The empirical case is developed elsewhere. In one sentence: the framework is a refinement of emotional-intelligence theory, showing that affective perceptiveness is not inherently adaptive unless coupled with calibrated situated attribution.

Podobné jednotky