The purpose of my research was to examine the impact of likability between two previously unacquainted individuals on memory conformity. One hundred and twenty seven undergraduate students were assigned to a likability condition (control, likable, or dislikable). After the likability manipulation the pair viewed pictures and were later tested on their memory for those pictures. The research confederate always answered first, so the participant’s responses could be based on a combination of their memory for the event and what the confederate reported. Results indicated that participants were most likely to conform to the confederate’s responses when in the dislikable condition. Participants were most likely to answer accurately when in the likable condition. Although contrary to the previous research examining memory conformity among friends and romantic partners, these results are supported by the rapport-building literature in which creating a comfortable relationship between two strangers results in more accurate recall.

Malgré les avancées considérables dans l’analyse des preuves matérielles, et le développement ces dernières années de la police scientifique, les témoignages oculaires occupent encore aujourd’hui un rôle primordial dans les décisions de justice. Et pourtant, ces témoignages sont faillibles. En effet, de nombreux facteurs se trouvant au croisement de processus mnésiques et/ou cognitifs et de processus sociaux et/ou sociocognitifs, peuvent les impacter de façon irréversible. C’est dans ce contexte, et dans le but de mieux comprendre certaines variables à l’origine de leur fragilité, que nous avons réalisé les sept expérimentations présentées dans cette thèse.De façon plus précise, les trois premières études s’intéressaient à une technique d’audition efficace, appelée l’entretien cognitif. Notre objectif était alors de développer et d’évaluer, en laboratoire et sur le terrain, un protocole modifié d’entretien cognitif fondé sur le principe de multiplication des rappels libres. Toutefois, dans cette thèse, nous souhaitions adopter une approche plus dynamique et situationnelle de l’entretien cognitif que celle rencontrée jusqu’à présent dans la littérature. C’est pourquoi, dans une seconde partie, nous avons examiné l’efficacité de ce protocole en lien avec trois variables évaluatrices indissociables des situations d’auditions, et pouvant dans les faits impacter fortement et négativement la qualité des témoignages oculaires. Ces trois variables sont: (a) les scripts que partagent les individus à propos des évènements criminels, (b) les discussions entre témoins, et, (c) les stéréotypes associés aux témoins par le biais de leursappartenances groupales.Plusieurs résultats ont alors été montrés. Tout d’abord, nous avons confirmé l’efficacité d’une version modifiée d’entretien cognitif (ECM). Plus précisément, un protocole composé de deux rappels libres, incluant les consignes d’exhaustivité et de remise en contexte ainsi qu’une nouvelle technique destinée à favoriser le souvenir, la focalisation périphérique guidée, a permis d’améliorer, dans toutes nos études, la richesse du rappel des participants sans nuire à son exactitude. Son efficacité a d’ailleurs été montrée aussi bien en laboratoire que sur le terrain. De plus, ce protocole intègre les consignes cognitives les plus efficaces et omet les moins effectives. Parallèlement, nous avons confirmé l’impact néfaste des scripts et des discussions entre témoins sur les témoignages oculaires. Certains effets des stéréotypes liés à l’appartenance groupale du témoin ont aussi été suggérés. Enfin, concernant l’efficacité de l’entretien cognitif, et plus précisément de la versionmodifiée, certains effets délétères de ce protocole et des consignes qui le composent ont été observés en lien avec les trois variables évaluatrices considérées. Toutefois, plusieurs bénéfices intéressants ont également été révélés dans ce cadre.Ces résultats seront discutés au regard des données disponibles dans la littérature à ce jour. Des recommandations appliquées seront également émises. / Despite many advances in analyzing physical evidence, and the development these past years of the forensic police, eyewitnesses’ testimonies remains decisive in the decisions of justice. Nevertheless, these testimonies are fallible. Numerous factors, at the crossroad of memory and/or cognitive processes and of social and/or sociocognitive processes, can impact them in an irreversible manner. In this context, we realized the seven experimentations presented in this dissertation. The aim was to understand better some variables responsible for the fragility of eyewitnesses’ accounts.More precisely, the first three studies presented in the first part of this work were dealing with an effective technique for interviewing eyewitnesses, called the Cognitive Interview. Our purpose was to develop and evaluate, in the lab and in the field, a modified version of Cognitive Interview, based on the principle of multiplication of the free recalls. However, in this dissertation, we chose to adopt a more dynamic and situational approach that the one encountered in the literature until now. Therefore, in a second part, we examined the efficacy of this protocol in interaction with three estimator variables, inseparable from the context of hearing witnesses, and which can in the real life strongly and negatively impact the quality of their accounts. These variables are: (a) the scripts shared by individuals about criminal events, (b) the talk between witnesses, and, (c) the stereotypes associated with witnesses through the social groups they belong to.Several results were shown. First, we confirmed the efficacy of a modified version of the Cognitive Interview (MCI). More precisely, a protocol composed of two free recallattempts, composed of the report all and the context reinstatement instructions, as well as a new technique designed to enhance memories, the guided peripheral focus, increased in all the studies the richness of participants’ recalls without impairing their accuracy. Its efficacy was shown in the lab and in the field. Moreover, this protocol includes the most effective cognitive instructions and omits the less beneficial. Then, in the second part, we confirmed that the scripts and the talk among witnesses have a detrimental impact on eyewitnesses’ testimonies. Some effects of the stereotypes linked to the group membership of the witness were also suggested. Finally, concerning the efficacy of the Cognitive Interview, and more precisely the modified protocol, some negative effects were observed in interaction with the three estimator variables considered. However, some interesting benefits of this protocol and of the cognitive instructions it includes were also shown.These results will be discussed in regards with the literature available today. Some applied recommendations will also be emitted.

Memory conformity effects occur when witnesses report misleading suggestions they learned about from another witness. Using a new paradigm the present thesis investigated whether what subject-witnesses report about an event also implies what they personally remember or know about that event. Subjects were tested in pairs, with each member of a pair shown a different version of a video using the MORI technique. There were critical details (e.g., theft of an iPod) in each of the following conditions: visible to only one member of each subject spair, visible to both members of the pair, and not visible to either member of the pair. Pairs subsequently completed a questionnaire together to remember details from the video. Subjects then individually completed a similar questionnaire. A source monitoring and subjective experiences test revealed that co-witness discussion does not necessarily lead people to experience illusory recollections for details they did not witness themselves.