Chatterbox Challenge as a Test-Bed for Synthetic Emotions

Chatterbox Challenge as a Test-Bed for Synthetic Emotions

Jordi Vallverdú (Universitat Autònoma de Barcelona, Spain), Huma Shah (Universitat Autònoma de Barcelona, Spain) and David Casacuberta (Universitat Autònoma de Barcelona, Spain)
Copyright: © 2010 |Pages: 26
DOI: 10.4018/jse.2010070102
OnDemand PDF Download:
No Current Special Offers


Chatterbox Challenge is an annual web-based contest for artificial conversational systems, ACE. The 2010 instantiation was the tenth consecutive contest held between March and June in the 60th year following the publication of Alan Turing’s influential disquisition ‘computing machinery and intelligence’. Loosely based on Turing’s viva voca interrogator-hidden witness imitation game, a thought experiment to ascertain a machine’s capacity to respond satisfactorily to unrestricted questions, the contest provides a platform for technology comparison and evaluation. This paper provides an insight into emotion content in the entries since the 2005 Chatterbox Challenge. The authors find that synthetic textual systems, none of which are backed by academic or industry funding, are, on the whole and more than half a century since Weizenbaum’s natural language understanding experiment, little further than Eliza in terms of expressing emotion in dialogue. This may be a failure on the part of the academic AI community for ignoring the Turing test as an engineering challenge.
Article Preview


In his anticipation of objections to the idea of machines thinking, and testing for it through an imitation game, Alan Turing reminded of a real-life scenario the viva voca in which an interrogator seeks answers to questions from a ‘witness’ (Turing, 1950). Pre-empting the argument from consciousness and quoting from Jefferson’s 1949 Lister Oration, “not until a machine can write a sonnet or compose a concerto because of thoughts and emotions felt ... not only write it but know that it had written it” (1950, section 6, p. 445), Turing countered showing this stance was a solipsistic one. To say that “no mechanism ... could feel pleasure at its successes, grief when its valves fuse, be warmed by flattery, be made miserable by its mistakes, be charmed by sex, be angry or depressed when it cannot get what it wants” (p. 446), was, according to Turing, an extreme position: “the only way by which one could be sure that a machine thinks is to be the machine and to feel oneself thinking” - Turing’s emphasis (ibid). Turing replied, as Stins and Laureys put it “in a succinct British fashion” (2009, p. 265) that rather than labouring over the point “A is liable to believe A thinks but B does not while B believes B thinks but A does not” it is “usual to have the polite convention that everyone thinks” (1950, p. 446).

There are a variety of objections to the “Turing test” and whether it can really be a way to assert whether a computer is thinking or not, it is beyond the scope of this paper to review them, readers are directed to Shah and Warwick (2010b) nevertheless the concept of being able to pass the Turing test is a complex endeavour indeed, and analysing such a variant Turing test contest can be of great philosophical value.

Turing put forward possible questions and answers to show that if the responses were “satisfactory and sustained” then one might not describe the answers as “an easy contrivance” (1950, p. 447). It is this method to assess whether machine responses to questions are satisfactory and sustained that is evaluated in an annual contest – the Chatterbox Challenge. Artificial conversation systems (ACE), commonly known as ‘chatbots’ compete against each other across a number of categories. What is considered a satisfactory response can be subjective; one interrogator may find a response inappropriate while another may accept it as humorous. An answer to a question may seem ersatz-like, but may also be a satisfactory emotive response under interrogation or conveying disinterest in topic by the ‘witness’. Whether machines are capable of expressing emotion through their responses or whether they are still Eliza-like (Weizenbaum, 1966) can be found by analysing the contest’s transcripts.

Complete Article List

Search this Journal:
Open Access Articles
Volume 13: 2 Issues (2022): Forthcoming, Available for Pre-Order
Volume 12: 2 Issues (2021): Forthcoming, Available for Pre-Order
Volume 11: 2 Issues (2020)
Volume 10: 2 Issues (2019)
Volume 9: 2 Issues (2018)
Volume 8: 2 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2015)
Volume 5: 2 Issues (2014)
Volume 4: 2 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing