Ethical considerations on recall methods in violent conflict research

Published on 3 May 2023

Veronica Corbellini, IDS Alumni MA GAD 2021-2022

Conflicts significantly impact countries, populations and individuals. Investigating and understanding new forms of power relations and social orders emerging in these contexts is necessary to break the cycle of silence surrounding them and contribute to peacebuilding and conflict prevention. Recall methods – requesting individuals to retrieve and narrate events they directly experienced – are increasingly used in research on violent conflict to gather further insights. However, war and conflict are classified as high-magnitude traumatic events, potentially exposing individuals to severe and multiple traumas. Consequently, the use of such methodology comes with risks and ethical dilemmas, which need to be carefully considered by researchers working in this area.

Research with individuals who experienced traumas related to conflict and violent settings falls under sensitive research, as those involved are potentially at risk of harm. In recall methods specifically, individuals are at high risk of being triggered and re-experience associated emotions when requested to narrate associated events.

While conversations and reflections on methodologies and ethics have progressed, researchers often have a limited understanding of how trauma and memory interact. Understanding this dynamic is essential to comprehend the potential impact of methodologies on participants. Using the Dual Representation Theory (DRT), a psychological model theorised by Professor Chris Brewin on trauma cognitive elaboration, we can reflect on the impact of trauma on memory and narrations and the ethical implications of using such a methodology.

Recall methods

According to Dual Representation Theory after experiencing an event, individuals elaborate the information and create durable representations in memory of the episode. Along with specific details (such as time and space), representations contain the individuals’ physiological, motor and emotional responses, and the meaning attributed. In recall methods, participants are requested to retrieve and narrate those representations.

The methodology has been applied to quantitative approaches to understand the effects of violent conflicts such as in the Nigeria General Household Survey (GHS) Panel. Additionally, in qualitative approaches, such as ethnographic methods or interviews, we are able to obtain insights and micro-level information. Viterna, for example, analysed the mobilisation processes that led women in El Salvador to participate in guerrilla armies.

To narrate past events, a necessary condition is the individual being an “internal witness” in control of those representations, especially the emotional component. When exposed to traumas – “any experience that threatens the health, wellbeing or survival of an individual or community” – the individual’s ability to perceive, process and respond to is highly impacted and disrupted. Consequently, they lack control and power over those memories.

The Dual Representation Theory is a psychological cognitive model of elaborating and memorising traumatic events. The theory helps explain the phenomena observed in individuals exposed to traumas, as well as the symptomatology associated with Post-Traumatic Stress Disorder (PTSD) – a psychiatric disorder that may occur in those who experienced or witnessed traumatic events. Even if not tested yet in violent and conflict contexts, it can provide several insights into social sciences research and practice.

Cognitive elaboration of traumatic events

The model theorises two cognitive representations. The verbally accessible memories (VAM) – or contextual representation (C-Rep)  are verbal and conscious, containing detailed and processed information about the event. Memories are retrieved automatically or through conscious strategies, and the individual can narrate them. The situationally accessible memories (SAM) – or sensory-bound representation (S-Rep) instead are non-verbal and unconscious. The sensory details, the individual’s emotional state and body response are stored without being understood and elaborated. Given their non-verbal nature, the individual struggles to retrieve and narrate them.

According to the model, in traumatic events, the sensory representation (S-rep) is strongly codified due to the high emotional relevance of the episode, while the contextual representation is either weakly encoded or not associated with the sensory one. Consequently, S-rep might be activated unconsciously or if the context presents meanings or features similar to the traumatic event, with the consequent emergence of the emotions experienced during the trauma – primarily fear, helplessness and horror.

Reflecting on outcomes

In discussing the ethicality of recall methods, two main areas are analysed – the selection process of participants and the interview outcomes:

Participants selectionFollowing the ‘do no harm’ principle, ethical concerns are directed at protecting subjects from any potential harm caused by participating in research. As a result, trauma-informed approaches have been increasingly used in research protocols, focusing on the selection process, interview structure, data collection and protection. Furthermore, as stated in the informed consent, participants are given the options of refusing to participate, declining to answer some or all the questions, or withdrawing from the study entirely.

However, two issues still emerge in adopting these precautions in research design. Firstly, informing the interviewee that the interview might be distressing does not necessarily deter individuals from participating in the research. On the contrary, they might be motivated to share their experience and insights to help researchers to understand the contexts better, contribute to the development of knowledge, as well as helping others.

Secondly, any participant can be carrying unresolved traumas. While questions might be formulated to avoid the complete recall or narration of a potentially traumatic event, researchers will always incur the risk of “tapping into unknown (or unknowable) sensitivities”. As theorised in the DRT model, questions or cues may unconsciously lead ”people to remember their trauma and losses and unless managed properly [they] could easily open up old wounds”.

Interview outcomes – As discussed earlier, the narration of events requires the individual to be an “internal witness” in control of the memory – including the related emotions. However, as theorised by Brewin, the memory of traumatised people is disorganised, presenting gaps in the recall process and difficulties in producing a coherent narration due to the pervasiveness of the associated emotions.

The outcomes of incompletely elaborated memories are various. The activation of the S-Rep might determine the uncontrolled emergence of emotional states experienced during the trauma. Alternatively, given that the S-Rep lacks verbal coding, individuals might be unable to narrate the event. In other cases, participants may be able to narrate the story in detail, but with an apparent lack of emotional response to the events. The individual might also be emotionally upset but have no control over their memory of the events, with even the possibility of being unable to recall them.

Towards wellbeing

When researching traumatic environments, it is vital to be mindful of the methodologies used while keeping the research subject’s wellbeing at the forefront of our considerations.

Firstly, attention should be directed to how interviews are conducted. Researchers should be trained in relevant psychological tools and in how to support the traumatic narrations of participants and consequent outcomes appropriately. The presence of psychologists and/or psychotherapists in the interview is also beneficial. However, the lack of such professional figures in these contexts is well acknowledged, in addition to concerns about applying a Western-centric discipline in non-Western settings.

Secondly, participants should receive additional support and space in the post-interview phase for elaboration. The elaboration will allow them not to be at the mercy of emotions and memories, or be left in distress, and ultimately restore a sense of control, safety and power while reducing future possible adverse reactions. At the same time, we need to be sensitive to the culture’s preferred method of elaborating traumas, which might not necessarily have a Western cultural preference for emotionally ventilating and working through what happened.

Ethical research goes beyond conforming to guidelines and protocols. It requires assuming a human, empathic and conscious position towards participants, researchers and the work itself.


The views expressed in this opinion piece are those of the author/s and do not necessarily reflect the views or policies of IDS.

Related content