Journal of Economic Behavior & Organization 106 (2014) 281–297 Contents lists available at ScienceDirect Journal of Economic Behavior & Organization j ourna l h om epa ge: w ww.elsevier.com/locate/jebo Signs and (counter)signals of trustworthiness  Diego Gambetta a,b , Áron Székely c,d,∗ a European University Institute, Via dei Roccettini 9, San Domenico, Firenze, Italy b Nuffield College, New Road, Oxford OX1 1NF, UK c University of Oxford, Department of Sociology, Manor Road, Oxford OX1 3UQ, UK d Institute of Cognitive Sciences and Technologies, CNR, Via San Martino della Battaglia 44, Rome, Italy a r t i c l e i n f o Article history: Received 4 October 2013 Received in revised form 17 July 2014 Accepted 20 July 2014 Available online 28 July 2014 JEL classification: D83 C72 C91 Keywords: Information Costly signals Generosity Trust Laboratory experiments a b s t r a c t When we become aware that our past actions carry information about qualities that we possess or lack, which others use to decide how to deal with us, are we unconcerned, content to rely on what we have done, or do we take action to alter this information? We study this question experimentally using generosity as a sign and a signal of trustworthiness, and a trust game. Subjects play a dictator game unaware that later they will play a trust game and that their level of generosity in the dictator game will be revealed to trusters, with some inaccuracy, before trusters decide whether to trust or not. Once made aware of what follows, trustees have the option to play a second dictator game, from which their choice will be accurately conveyed to trusters in addition to their decision in the initial game. Consistent with ‘countersignalling theory’, those who, in the first dictator game, were either miserly or generous do not play the second dictator game, resigned or content with the information conveyed by their past actions. Those neither miserly nor generous in the first dictator game, an intermediate generous group, are likeliest to use the second dictator game; many of them for the purpose of signalling, so that they are not confused with the miserly. © 2014 Elsevier B.V. All rights reserved. 1. Introduction We investigate a common case in which people realise that some past action of theirs, which they performed for motives other than informing others of their qualities, becomes known to others in situations in which this could have consequences for them. Going about their lives, people act in myriad ways, good and bad. Many of these actions are trivial and go unnoticed, forever lost. But some actions are perceived by other people, other actions are recorded on media whether agents are aware of it or not, others still leave enduring marks on them, perceivable well after the event that produced them occurred. The information potential contained in these past actions can remain dormant. However, out of this baggage with which everyone travels, people often extract signs of other people’s qualities, good or bad, and use this information in deciding how to deal with them.  We thank Ozan Aksoy, Dominic Burbidge, Giorgia Gon, Dan Honig, Wojtek Przepiorka, Dingeman Wiertz, Nan Zhang, and two anonymous referees for their helpful comments. This work was supported by the Economic and Social Research Council [Grant number ES/I022066/1]. Data are available upon request from the authors. ∗ Corresponding author at: Institute of Cognitive Sciences and Technologies, CNR, Via San Martino della Battaglia 44, Rome, Italy. Tel.: +39 06 44362370x1; fax: +39 06445 952 43. E-mail addresses: diego.gambetta@eui.eu (D. Gambetta), aron.szekely@sociology.ox.ac.uk (Á. Székely). http://dx.doi.org/10.1016/j.jebo.2014.07.009 0167-2681/© 2014 Elsevier B.V. All rights reserved.