The manic phase of Bipolar disorder significantly impairs theory of mind decoding
Author links open overlay panelEmily R. Hawken a b 1, Kate L. Harkness b c 1, Lauren K. Lazowski a b c d , David Summers a b c d , Nida Khoja a b c d , James Gardner Gregory d , Roumen Milev b c d
Cite https://doi.org/10.1016/j.psychres.2016.03.043 Get rights and content Highlights • Manic phase was significantly associated with decreased accuracy at decoding mental states in the Eyes task. • Eyes task performance was negatively correlated with the symptoms of language/thought disorder, pressured speech, and disorganized thoughts and appearance. • Findings held when controlling for accuracy on the Animals task, response times, and relevant demographic and clinical covariates. Abstract Bipolar disorder is associated with significant deficits in the decoding of others’ mental states in comparison to healthy participants. However, differences in theory of mind decoding ability among patients in manic, depressed, and euthymic phases of bipolar disorder is currently unknown. Fifty-nine patients with bipolar I or II disorder (13 manic, 25 depressed, 20 euthymic) completed the “Reading the Mind in the Eyes” Task (Eyes task) and the Animals Task developed to control for non-mentalistic response demands of the Eyes Task. Patients also completed self-report and clinician-rated measures of depression, mania, and anxiety symptoms. Patients in the manic phase were significantly less accurate than those in the depressed and euthymic phases at decoding mental states in the Eyes task, and this effect was strongest for eyes of a positive or neutral valence. Further Eyes task performance was negatively correlated with the symptoms of language/thought disorder, pressured speech, and disorganized thoughts and appearance. These effects held when controlling for accuracy on the Animals task, response times, and relevant demographic and clinical covariates. Results suggest that the state of mania, and particularly psychotic symptoms that may overlap with the schizophrenia spectrum, are most strongly related to social cognitive deficits in bipolar disorder. Introduction Bipolar disorder is associated with marked deficits in social and interpersonal functioning that persist into euthymia (Sanchez-Moreno et al., 2009). Because the dysfunction associated with bipolar disorder can be so profound, it is important examine the social-cognitive mechanisms that might underlie these problems. Theory of mind (ToM) – the ability to decode and reason about others’ mental states, including beliefs, desires, emotions, and intentions – forms the foundation of social cognition and, thus, is critical to successful social and interpersonal functioning (Premack and Woodruff, 1978). There are two components of theory of mind that are distinct, but that work together to facilitate social understanding: 1) decoding mental states from immediately available social information (e.g., facial expression, tone of voice), and 2) reasoning about mental states by using knowledge about others’ experiences and beliefs to understand behavior (Sabbagh, 2004). This conceptualization of ToM, and particularly the construct of ToM reasoning, is complementary to the more recent distinction between ‘ToM-understanding’ and ‘ToM-use’, which have been defined as the ability to understand others’ mental states and to apply that understanding in social situations, respectively (Abu-Akel, 2003, Wang et al., 2013, Wang et al., 2015). Theory of mind ‘decoding’ is theorized to represent the foundational component of ToM (Sabbagh, 2004) and is the component that we sought to examine in the current study. The “Reading the mind in the eyes task” (Eyes Task; Baron-Cohen et al., 2001) is the most widely used test of theory of mind decoding in adults. This difficult task requires individuals to judge the complex mental states portrayed in pictures of the eye region of faces using a forced choice among four mental state adjectives (e.g., reflective, interested, flirtatious, bored). As such it is capable of detecting very subtle differences in social intelligence. Deficits in ToM decoding using the Eyes task have been observed across a range of psychiatric conditions associated with social and interpersonal dysfunction, including autism spectrum disorder (e.g., Baron-Cohen et al., 2001), schizophrenia (e.g., Sprong et al., 2007), and unipolar major depressive disorder (e.g., Lee et al., 2005). A recent meta-analysis of 12 studies that compared patients with bipolar disorder to healthy controls using the Eyes task found a significant effect size (d=.50; Bora et al., 2016). All but one of these studies included bipolar disorder patients in the euthymic phase or with subsyndromal manic or depression symptoms. The one study that included patients in the acute phases of illness also found significant impairment in the patient group relative to healthy controls (Wiener et al., 2011). Further, overall effect sizes across all ToM tasks in the full set of 34 studies included in the meta-analysis were most robust for differences between healthy controls and those in acute manic or depressive states (d=1.32). While differences between patients with bipolar disorder and healthy controls on ToM tasks generally, and the Eyes task in particular, are robust and well-documented, to date there has been no comparison of ToM decoding abilities across the manic, depressed, and euthymic phases of bipolar disorder. This is important because there are reasons to suspect that individuals in the manic phase of bipolar disorder may be particularly impaired in their ToM decoding skills relative to those in the depressed or euthymic phases. First, in studies of ToM reasoning, patients in the manic phase perform worse than those in the depressed phase of illness (Kerr et al., 2003), although this group difference has failed to reach significance in other studies using a variety of different ToM reasoning measures (Bazin et al., 2009, Wolf et al., 2010). Second, the manic phase of bipolar disorder shares some features with the schizophrenia spectrum, including language/thought disorder (e.g., pressured speech, tangentiality, flight of ideas) and delusions of reference/grandeur, and molecular and behavioral genetic work suggests that bipolar disorder may, along with schizophrenia, form part of a spectrum of neurodevelopmental disorders (Lichtenstein et al., 2009; Van Snellenberg and de Candia, 2009). In schizophrenia, deficits in theory of mind have been specifically linked to formal thought disorder (Greig et al., 2004), and researchers have suggested that poor ‘mind-reading’ (i.e. difficulty decoding and reasoning about others’ mental states) may actually cause pragmatic impairments in thought-language expression and comprehension (Langdon et al., 2002). Similarly, Frith, 1992, Frith, 1994 noted that patients with delusions of reference show marked ToM deficits, and that their delusions occur due to a lack of ability to represent others’ beliefs, emotions, and intentions. Lahera et al. (2008), however, found that a previous history of psychotic symptoms was not associated with pronounced ToM deficits in bipolar euthymic patients. Furthermore, Bora et al. (2016) did not find a significant difference in ToM deficits between bipolar disorder patients with or without a history of psychosis, leading authors to question whether or not ToM deficits are a trait-marker for psychosis (Wang et al., 2008, Mitchell and Young, 2016). However, in these two latter studies patients were either examined in the euthymic phase, or phase of illness was not taken into consideration. To the extent that patients in the manic phase of bipolar disorder exhibit similar symptoms to schizophrenia, we may expect similar theory of mind decoding deficits. Indeed, in a meta-analysis examining several ToM tasks, severity of manic symptoms was associated with the degree of ToM performance (Bora et al., 2016). The primary goal of the current study was to compare ToM decoding abilities assessed with the Eyes task among patients in the manic, depressed, or euthymic states of bipolar I or II disorder. We hypothesized that patients in the manic phase would perform significantly worse than those in the depressed and euthymic phases. In contrast, consistent with previous research, we did not expect to see differences in performance between those in the depressed and euthymic phases. Further, we predicted that the deficits in performance associated with the manic phase would be specifically driven by symptoms indicating thought/language disorder and delusions. Finally, we hypothesized that the above effects would be robust to individual differences across groups in response times associated with the task and in performance on a non-mentalistic control task.
2016年に発表されたEmily R. Hawkenらによる論文『The manic phase of Bipolar disorder significantly impairs theory of mind decoding（双極性障害の躁状態は、心の理論における解読能力を著しく損なう）』の要約と詳しい解説をまとめます。
この研究は、双極性障害の各病相（躁、うつ、寛解）が、対人関係に不可欠な「他人の心を読み取る能力」にどのような影響を与えるかを比較した非常に重要な論文です。
論文要約
1. 背景と目的
双極性障害（BD）の患者は、症状が落ち着いている時期（寛解期）でも、対人関係や社会的機能に問題を抱えることが多いことが知られています。その原因の一つとして、「心の理論（Theory of Mind: ToM）」、つまり他人の意図や感情、信念を推測する能力の低下が疑われてきました。
本研究は、躁状態、うつ状態、寛解期の患者において、この「心を読み取る能力」がどのように異なるかを明らかにすることを目的としました。
2. 研究方法
- 対象者: 双極性障害患者 59名（躁状態13名、うつ状態25名、寛解期20名）。
- 評価課題:
- 「目の課題（Eyes Task）」: 写真に写った「目元」だけを見て、その人が何を考えているか（複雑な心理状態）を4つの選択肢から選ぶテスト。
- 「動物課題（Animals Task）」: 目の課題と同じ難易度で、感情ではなく単に動物の種類などを判別するテスト（認知能力全般の低下を区別するための対照実験）。
- 臨床評価: 思考障害、談話促迫（しゃべり続け）、身なりの乱れなどの症状も測定。
3. 主な結果
- 躁状態の圧倒的な成績低下: 躁状態の患者は、うつ状態や寛解期の患者に比べて、他人の心理状態を読み取る正確さが有意に低いことが判明しました。
- 特に「ポジティブ・中立」な表情に弱い: 躁状態の患者は、相手が好意的であったり、中立的な表情をしていたりする場合、その意図を読み間違える傾向が最も強く見られました。
- 思考の乱れとの相関: 心理状態の読み取りミスは、「談話促迫（猛烈にしゃべる）」「思考の混乱」「身なりの乱れ」といった症状が強いほど深刻でした。
- うつ・寛解期では差がない: 意外なことに、うつ状態の患者と寛解期の患者の間では、この能力に大きな差は見られませんでした。
4. 結論
双極性障害における社会的な認知機能の低下は、主に躁状態によって引き起こされます。特に躁状態で見られる「思考の乱れ（統合失調症と重なる症状）」が、他人の心を正しくデコード（解読）する能力を著しく妨げていることが示唆されました。
詳しく解説：なぜ「躁状態で心が読めなくなる」のか？
この論文が持つ重要な臨床的意味を3つのポイントで解説します。
① 「躁」は単なる「元気」ではない
一般的に躁状態は「気分が良くて社交的」と思われがちですが、実際には「一方的なコミュニケーション」に陥りやすいのが特徴です。
この研究は、その理由を解明しています。躁状態では脳の興奮が強すぎて、相手の目元にある「繊細なサイン（ためらい、興味、困惑など）」を拾い上げる情報の解読プロセス（デコーディング）が壊れてしまっているのです。
② 統合失調症スペクトラムとの重なり
論文は、躁状態で見られる「思考障害（話が飛ぶ、理屈が通らない）」が、心の理論の欠如と深く関わっていることを指摘しています。
これは、以前議論した「神経免疫（ミクログリアの暴走）」や「ミトコンドリアの異常」によって、脳の高度な情報処理ネットワークが一時的にショートし、「自分と他人の境界線」や「他者の視点」を処理できなくなっている状態を反映していると考えられます。
③ 対人関係の悪化と「躁」の結びつき
躁状態の患者が周囲とトラブルを起こしやすいのは、単に「わがまま」だからではなく、「相手が困っている、あるいは嫌がっているという非言語的なサインを、脳が物理的に処理できていない」ためであると言えます。
特にポジティブや中立的な表情の読み取りに失敗するという結果は、「相手が普通に接していても、それを過剰にポジティブに誤解したり、あるいは意図を無視したりする」メカニズムを説明しています。
過去の議論（躁病先行仮説）との関連性
この論文の結果は、クコプロスの「躁病先行仮説」を補完するものです。
もし「躁こそが疾患の主導的なエンジン」であるならば、双極性障害特有の「社会生活の破綻」や「対人関係の崩壊」もまた、躁状態（あるいは軽躁状態）における認知のバグから始まっていることになります。
まとめると：
この論文は、「躁状態は、脳の社会的な受信アンテナを麻痺させる」ことを明らかにしました。うつ状態よりも躁状態において「相手の心が読めなくなる」という事実は、双極性障害の治療において躁のコントロールがいかに最優先であるかを、対人関係の観点からも裏付けています。
