Abstract
To be really effective, conversational agents must integrate well with the characteristics of the humans with whom they interact. This exploratory study focuses on a method for integrating well-assessed methods from the field of social psychology in the design of task-oriented conversational agents in which the dialogue management module is developed through machine learning. In particular, the aim is to achieve agents whose policies could take into account the psychological features of the human interactants to deliver personalized and more effective messages. The paper presents the psychological study performed and outlines the overall theoretical architecture of the software framework proposed. On the psychosocial side, we first assessed the effectiveness of differently framed messages aimed to reducing red meat consumption taking the Theory of Planned Behavior (TPB) as the psychosocial model of reference. Turning to the machine learning field, the resulting Structural Equation Model (SEM) was first translated into a probabilistic predictor using Dynamic Bayesian Network (DBN). In turn, such DBN became the fundamental element of a Partially Observable Markov Decision Processes (POMDP) in a reinforcement learning setting. The possibility to elicit complete interaction policies was then studied by applying Neural Monte Carlo Tree Search (Neural MCTS) methods. The results thus obtained introduce the possibility to develop new multidisciplinary and integrated techniques for the development of automated dialogue managing systems.
Similar content being viewed by others
References
Ajzen I (1991) The theory of planned behavior. Organ Behav Hum Decis Process 50(2):179–211. https://doi.org/10.1016/0749-5978(91)90020-t
Allen J, Ferguson G, Stent A (2001) An architecture for more realistic conversational systems. In: Proceedings of the 6th international conference on intelligent user interfaces. https://doi.org/10.1145/359784.359822. ACM, pp 1–8
Bach-Faig A, Berry EM, Lairon D, Reguant J, Trichopoulou A, Dernini S, Medina FX, Battino M, Belahsen R, Miranda G, Serra-Majem L (2011) Mediterranean diet pyramid today. Sci Cult Updates Public Health Nutr 14(12A):2274–2284. https://doi.org/10.1017/s1368980011002515
Ben GI (2007) Bayesian networks. encyclopedia of statistics in quality and reliability. John Wiley & Sons, New York
Bertolotti M, Carfora V, Catellani P (2020) Regulatory focus and the effect of nutritional messages on health and well-being: the case of red meat intake. Appl Psychol Health Well Being 12(1):212–230
Bertolotti M, Carfora V, Catellani P (2020) Different frames to reduce red meat intake: the moderating role of self-efficacy. Health Commun 35 (4):475–482. https://doi.org/10.1111/aphw.12180
Bianchi F, Aveyard P, Astbury NM, Cook B, Cartwright E, Jebb SA (2019) Replacing meat with alternative plant-based products (RE-MAPs): protocol for a randomised controlled trial of a behavioural intervention to reduce meat consumption. BMJ open 9(5):e027016. https://doi.org/10.1136/bmjopen-2018-027016
Bohm I, Lindblom C, Åbacka G, Bengs C, Hörnell A (2015) “He just has to like ham” – The centrality of meat in home and consumer studies. Appetite 95:101–112. https://doi.org/10.1016/j.appet.2015.06.015
Bosone L, Martinez F (2017) When, how and why is loss-framing more effective than gain-and non-gain-framing in the promotion of detection behaviors?. Int Rev Soc Psychol 1(30):184–192. https://doi.org/10.5334/irsp.15
Carfora V, Bertolotti M, Catellani P (2019) Informational and emotional daily messages to reduce red and processed meat consumption. Appetite 141:104331. https://doi.org/10.1080/10410236.2019.1567444
Carfora V, Caso D, Conner M (2017) Randomised controlled trial of a text messaging intervention for reducing processed meat consumption: the mediating roles of anticipated regret and intention. Appetite 117:152–160. https://doi.org/10.1016/j.appet.2017.06.025
Carfora V, Caso D, Palumbo F, Conner M (2018) Promoting water intake. the persuasiveness of a messaging intervention based on anticipated negative affective reactions and self-monitoring. Appetite 130:236–246. https://doi.org/10.1016/j.appet.2018.08.017
Carfora V, Catellani P, Caso D, Conner M (2019) How to reduce red and processed meat consumption by daily text messages targeting environment or health benefits. J Environ Psychol 65:101319. https://doi.org/10.1016/j.jenvp.2019.101319
Caso D, Carfora V (2017) Messaging intervention for promoting self-monitoring of fruit and vegetable consumption — [Un intervento di messaggistica istantanea per la promozione del monitoraggio del consumo di frutta e verdure]. Psicologia della Salute 1:97–111. https://doi.org/10.3280/PDS2017-001005
Cesario J, Corker KS, Jelinek S (2013) A self-regulatory framework for message framing. J Exp Soc Psychol 49:238–249. https://doi.org/10.1016/j.jesp.2012.10.014
Chaslot G, Bakkes S, Szita I, Spronck P (2008) Monte-Carlo tree search: a new framework for game AI. Bijdragen
Cheng T, Woon DK, Lynes JK (2011) The use of message framing in the promotion of environmentally sustainable behaviors. Soc Mar Q 17 (2):48–62. https://doi.org/10.1080/15245004.2011.570859
Corrin T, Papadopoulos A (2017) Understanding the attitudes and perceptions of vegetarian and plant-based diets to shape future health promotion programs. Appetite 10:40–47. https://doi.org/10.1016/j.appet.2016.11.018
Dagum P, Galper A, Horvitz E (1992) Dynamic Network Models For Forecasting. In: Proceedings of the eighth conference on uncertainty in artificial intelligence
de Carolis B, Mazzotta I (2017) A user-adaptive persuasive system based on ‘a-rational’ theory. Int J Hum-Comput St 108:70–88. https://doi.org/10.1016/j.ijhcs.2017.05.005
Dijkstra A, Rothman A, Pietersma S (2011) The persuasive effects of framing messages on fruit and vegetable consumption according to regulatory focus theory. Psychol Health 26(8):1036–1048. https://doi.org/10.1080/08870446.2010.526715
Eagly AH, Chaiken S (1993) The psychology of attitudes. Harcourt Brace Jovanovich College Publishers. https://doi.org/10.1002/mar.4220120509
Eshel G, Martin PA (2006) Diet, energy, and global warming, Earth Interactions (10)1–17. https://doi.org/10.1175/ei167
Fabiani P, Teichteil-Königsbuch F (2010) Markov decision processes in artificial intelligence. Wiley-ISTE, New York
Farchi S, De Sario M, Lapucci E, Davoli M, Michelozzi P (2017) Meat consumption reduction in Italian regions: health co-benefits and decreases in GHG emissions. Plos One 12(8):e0182960. https://doi.org/10.1371/journal.pone.0182960
Godinho CA, Alvarez MJ, Lima ML (2016) Emphasizing the losses or the gains: comparing situational and individual moderators of framed messages to promote fruit and vegetable intake. Appetite 96:416–425. https://doi.org/10.1016/j.appet.2015.10.001
Graça J, Calheiros MM, Oliveira A (2015) Attached to meat? (Un)Willingness and intentions to adopt a more plant-based diet. Appetite 95:113–125. https://doi.org/10.1016/j.appet.2015.06.024
Hancock GR, Samuelsen KM (2007) Advances in latent variable mixture models IAP
Iacobucci D (2010) Structural equations modeling: Fit indices, sample size, and advanced topics. J Consum Psychol 20:90–98. https://doi.org/10.1016/j.jcps.2009.09.003
Kaelbling PL, Littman M, Anthony RC (1998) Planning and acting in partially observable stochastic domains. Artif Intell 101:99–134
Katt S, Oliehoek FA, Amato C (2017) Learning in POMDPs with monte carlo tree search proceedings of the 34th international conference on machine learning - Volume 70, ICML 2017
Kline R (1988) Principles and practice of structural equation modeling. Guilford Press, New York
Kocsis L, Szepesvári C (2006) Bandit based monte-carlo planning. Springer, Berlin, pp 282–293
Laczniak RN, Muehling DD (1993) Toward a better understanding of the role of advertising message involvement in ad processing. P&M 10(4):301–319. https://doi.org/10.1002/mar.4220100405
Liu B (2018) Learning task-oriented dialog with neural network methods. PhD thesis
Misra R, Balagopal P, Raj S, Patel TG (2018) Red meat consumption (heme iron intake) and risk for diabetes and comorbidities?. Curr Diab Rep 18 (11):100. https://doi.org/10.1007/s11892-018-1071-8
Pearl J (1988) Newblock Probabilistic Reasoning in Intelligent systems: Networks of Plausible Inference. Representation and Reasoning Series (2nd printing ed.) Morgan Kaufmann, San Francisco
Petty RE, Cacioppo JT (1986) The elaboration likelihood model of persuasion. In: Communication and persuasion. https://doi.org/10.1007/978-1-4612-4964-1_1. Springer, New York, pp 1–24
Petty RE, Cacioppo JT (2012) Communication and persuasion: central and peripheral routes to attitude change. Springer Science & Business Media, Berlin
Rothman AJ, Bartels RD, Wlaschin J, Salovey P (2006) The strategic use of gain-and loss-framed messages to promote healthy behavior: how theory can inform practice. J Commun 56(suppl_1):S202–S220. https://doi.org/10.1111/j.1460-2466.2006.00290.x
Silver D, Schrittwieser J, Simonyan K, et al. (2017) Mastering the game of go without human knowledge. Nature 550(7676):354–359. https://doi.org/10.1038/nature24270
Silver David, Veness J (2010) Monte-Carlo planning in large POMDPs in advances in neural information processing systems, 23(NIPS 2010) 2164–2172
Stea S, Pickering GL (2019) Optimizing messaging to reduce red meat consumption. Environ Commun 13(5):33–648. https://doi.org/10.1080/17524032.2017.1412994
Stephenson MT, Holbert RL (2003) A Monte Carlo simulation of observable versus latent variable structural equation modeling techniques. Commun Res 30(3):332–354. https://doi.org/10.1177/0093650203030003004
Sutton R, Barto GA (1998) Reinforcement learning: an introduction. IEEE transactions on neural networks / a publication of the IEEE Neural Networks Council 9:1054
Vainio A, Irz X, Hartikainen H (2018) How effective are messages and their characteristics in changing behavioural intentions to substitute plant-based foods for red meat? the mediating role of prior beliefs. Appetite 125:217–224. https://doi.org/10.1016/j.appet.2018.02.002
Weller KE, Greene GW, Redding CA, et al. (2014) Development and validation of green eating behaviors, stage of change, decisional balance, and self-efficacy scales in college students. J Nutr Educ Behav 46(5):324–333. https://doi.org/10.1016/j.jneb.2014.01.002
Wood W (2000) Attitude change: persuasion and social influence. Annu Review Psychol 51(1):539–570. https://doi.org/10.1146/annurev.psych.51.1.539
Zur I, Klöckner C A (2014) Individual motivations for limiting meat consumption. Br Food J 116(4):629642. https://doi.org/10.1108/BFJ-08-2012-0193
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Valentina Carfora and Francesca Di Massimo contributed equally to this work.
Appendix: Persuasive messages
Appendix: Persuasive messages
Gain | Non-loss | Non-gain | Loss |
---|---|---|---|
Messages | Messages | Messages | Messages |
If you eat little red meat and cold cuts, you will improve the health of your stomach. | If you eat little red meat and and cold cuts, you will avoid damaging the health of your stomach. | If you eat much red meat and and cold cuts, you will miss the chance to improve the health of your stomach. | If you eat much red meat and and cold cuts, you will damage the health of your stomach. |
If you eat little red meat and cold cuts, you will improve the functioning of your bowel. | If you eat little red meat and and cold cuts, you will avoid damaging the functioning of your bowel. | If you eat much red meat and and cold cuts, you will miss the opportu-nity to improve the functioning of your bowel. | If you eat much red meat and and cold cuts, you will damage the functioning of your bowel. |
If you eat little red meat and cold cuts, you will improve the functionality of your heart. | If you eat little red meat and and cold cuts , you will avoid damaging the functionality of your heart. | If you eat much red meat and and cold cuts, you will miss the chance to improve the functioning of your heart. | If you eat much red meat and and cold cuts, you will damage the functionality of your heart. |
If you eat little red meat and cold cuts, you will improve the proper functioning of your arteries. | If you eat little red meat and and cold cuts, you will avoid increasing the malfunctioning of your arteries. | If you eat much red meat and and cold cuts, you will miss the opportunity to improve the proper functioning of your arteries. | If you eat much red meat and and cold cuts, you will increase the malfunctioning of your arteries. |
If you eat little red meat and cold cuts, you will enhance the functionality of your kidneys. | If you eat little red meat and and cold cuts, you will avoid straining the functionality of your kidneys. | If you eat much red meat and and cold cuts, you will miss the chance to enhance the functionality of kidneys. | If you eat much red meat and and cold cuts, you will strain the functionality of your kidneys. |
If you eat little red meat and cold cuts, you will enhance the health of your lungs. | If you eat little red meat and and cold cuts, you will avoid damaging the health of your lungs. | If you eat much red meat and and cold cuts, you will miss the opportunity to enhance the health of your lungs. | If you eat much red meat and and cold cuts, you will damage the health of your lungs. |
If you eat little red meat and cold cuts, you will enhance the health of your pancreas. | If you eat little red meat and and cold cuts, you will avoid damaging the health of your pancreas. | If you eat much red meat and and cold cuts, you will miss the chance to enhance the health of your pancreas. | If you eat much red meat and and cold cuts, you will damage the health of your pancreas. |
If you eat little red meat and cold cuts, you will improve the chance of having an optimal blood pressure. | If you eat little red meat and and cold cuts, you will decrease the chance of having hypertension. | If you eat much red meat and and cold cuts, you will miss the chance of having an optimal blood pressure. | If you eat much red meat and and cold cuts, you will increase the chance of having hypertension. |
Rights and permissions
About this article
Cite this article
Carfora, V., Di Massimo, F., Rastelli, R. et al. Dialogue management in conversational agents through psychology of persuasion and machine learning. Multimed Tools Appl 79, 35949–35971 (2020). https://doi.org/10.1007/s11042-020-09178-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-020-09178-w