Authors: Takume Takehara, Fumio Ochiai, & Naoto Suzuki
Title: Scaling laws in emotion-associated words and corresponding network topology
Journal: Cognitive Processing
doi: 10.1007/s10339-014-0643-z
Abstract: We investigated whether scaling laws were present in the appearance-frequency distribution of emotion-
associated words and determined whether the network constructed from those words had small-world or scale-free 
properties. Over 1,400 participants were asked to write down the first single noun that came to mind in response to nine 
emotional cue words, resulting in a total of 12,556 responses. We identified Zipf’s law in the distribution of the data, 
as the slopes of the regression lines reached approximately -1.0 in the appearance frequencies for each emotional cue 
word. This suggested that the emotion-associated words had a clear regularity, were not randomly generated, were scale-
invariant, and were influenced by unification/diversification forces. Thus, we predicted that the emotional intensity of 
the words might play an important role for a Zipf’s law. Moreover, we also found that the 1-mode network of emotion-
associated words clearly had small-world properties in terms of the network topologies of clustering, average distance, 
and small-worldness value, indicating that all nodes (words) were highly interconnected with each other and were only a 
few short steps apart. Furthermore, the data suggested the possibility of a scale-free property. Interestingly, we were 
able to identify hub words with neutral emotional content, such as ‘dog’, ‘woman’, and ‘face’, indicating that these 
neutral words might be an intermediary between words with conflicting emotional valence. Additionally, efficiency and 
optimal navigation in terms of complex networks were discussed.
著者Contact先の email: takehara[at]mail.doshisha.ac.jp
日本語によるコメント: 本研究では、感情手がかり語から連想される感情連想語の出現頻度分布にスケーリング則が存在するかどう

Authors: Takuma Takehara, Fumio Ochiai, & Naoto Suzuki
Title: A small-world network model of facial emotion recognition
Journal: Quarterly Journal of Experimental Psychology
doi: 10.1080/17470218.2015.1086393
Abstract: Various models have been proposed to increase understanding of the cognitive basis of facial emotions. Despite 
those efforts, interactions between facial emotions have received minimal attention. If collective behaviours relating to 
each facial emotion in the comprehensive cognitive system could be assumed, specific facial emotion relationship patterns 
might emerge. In this study, we demonstrate that the frameworks of complex networks can effectively capture those 
patterns. We generate 81 facial emotion images (6 prototypes and 75 morphs) and then ask participants to rate degrees of 
similarity in 3240 facial emotion pairs in a paired comparison task. A facial emotion network constructed on the basis of 
similarity clearly forms a small-world network, which features an extremely short average network distance and close 
connectivity. Further, even if two facial emotions have opposing valences, they are connected within only two steps. In 
addition, we show that intermediary morphs are crucial for maintaining full network integration, whereas prototypes are 
not at all important. These results suggest the existence of collective behaviours in the cognitive systems of facial 
emotions and also describe why people can efficiently recognize facial emotions in terms of information transmission and 
propagation. For comparison, we construct three simulated networks?one based on the categorical model, one based on the 
dimensional model, and one random network. The results reveal that small-world connectivity in facial emotion networks is 
apparently different from those networks, suggesting that a small-world network is the most suitable model for capturing 
the cognitive basis of facial emotions.
著者Contact先の email: takehara[at]mail.doshisha.ac.jp
日本語によるコメント: カテゴリモデルや次元モデル等、従来の表情認知モデルは1つ1つの表情刺激同士の関連性や全体としての次