Together with Benjamin Jainta

When we remember personally experienced episodes, we mentally travel back in time and create a scenario of the past. The novel framework of our research group suggests that during scenario construction, the episodic memory trace, entailing information about an episodes sequential structure, is enriched with relevant semantic information. However, episodic memories are not like exact video-recordings of our past that we can simply watch over and over again. On the contrary: episodic memories are flexible and prone to change. It is believed that internal models derived from memories are regularly updated based on new information which allows us to maintain valuable predictions in an ever-changing environment. What still remains to be understood is (1) how the two components in scenario construction (sequential and semantic information) are involved in memory updating on a neural level and (2) which conditions render episodic memories more or less prone to change. To address these questions, we conducted a functional magnetic resonance imaging (fMRI) experiment in which we violated expectations based on before encoded episodes on a sequential and semantic level and additionally applied two experimental factors, self-performance and self-perspective, which manipulated the experiential quality of episodes. In our talk, we will focus on presenting the methods of our study which will also serve as a basis for our follow-up experiments and discuss preliminary results.

Zoom meeting room details:

Link: 

 https://ruhr-uni-bochum.zoom.us/j/99840180341?pwd=anNRSk9WakRLd01tR1ZuMnBiaUlwQT09

Meeting ID: 998 4018 0341

Passcode:  Fb852v

Memory research has accumulated a wealth of data over the last decades. However, it seems that the more we learn, the less we know about what memory is and how it works. I propose that the way forward is to acknowledge that memory is always studied indirectly and that observed behavior is always the result of a complex interaction of several cognitive processes, which in isolation can be fairly simple. By contrast, the conventional interpretation is that any observed behavioral effect in a memory study is the result of memory processing, which makes memory seem highly complex. Taking a computational model of recognition memory as an example, I will show that many factors that previously were interpreted as attributes of the recognition memory system can be modeled as the result of other factors such as the stimulus statistics or a decision making process. Even the need for two distinct memory processes, called recollection and familiarity, vanishes in our model. This view of a simple generic memory that is used in a much more complex process has enabled us to account for other phenomena such as ABA renewal and the dependence of visual discrimination on the hippocampus. We believe that this approach has even wider applicability and work towards extending it.

Zoom meeting room details:

Link: 

 https://ruhr-uni-bochum.zoom.us/j/99840180341?pwd=anNRSk9WakRLd01tR1ZuMnBiaUlwQT09

Meeting ID: 998 4018 0341

Passcode:  Fb852v

Episodic memory can be thought of as involving a form of temporal perspective-taking. The remember imaginatively occupies a point in time in the past. This is also sometimes referred to as mental time travel. I will offer some reflections on what this human ability to engage in temporal perspective-taking consists in, including how it differs from spatial perspective-taking and how it might be related to counterfactual reasoning abilities.  

Zoom meeting room details:

Link: 

 https://ruhr-uni-bochum.zoom.us/j/93607143813?pwd=d1A1QVhydTFkNnpHK2xrWHh2MEg1dz09

Meeting ID: 936 0714 3813

Passcode:  374170

Zoom meeting room details:

Link: 

https://ruhr-uni-bochum.zoom.us/j/97457942133?pwd=SGVEdTZmY0UxZGEvVlkzZlYyZnpxUT09

Meeting ID: 974 5794 2133

Passcode:  374170

Judgments based on episodic memory are often thought to be immune to errors of misidentification (IEM). Yet there is a certain category of episodic memories that seems to threaten this supposed immunity of our memory-based judgments. Are observer memories a threat to the (potential) IEM feature of memory-based judgments? In recent years the phenomenon of observer memories has raised a growing amount of interest amongst philosophers, in particular due to its potential implication for the issue of IEM. In the resulting ongoing debate, some say that observer memories are a threat to the IEM enjoyed by episodic memory (Michaelian, 2020); others say that such memories pose no such threat (Fernández, 2018; Lin, 2020). In this paper, we argue for a middle way. First, we frame the debate, claiming that the existing literature lacks a satisfying definition both of observer memories and of the precise issue of errors of identification in such memories. Then, we contribute to the debate by challenging an anti-separatist assumption about the relation between phenomenal and intentional features of observer memories that looms behind this debate. On this assumption, if the rememberer’s self is built into the phenomenal content, by implication it is also built into the intentional content. We reject this assumption and offer a moderate separatist account of the relation between phenomenal and intentional content. Then, distinguishing between different species of observer memories by drawing upon the empirical literature, we say that for one species the rememberer’s self is part of the phenomenal content but not built into the intentional content: such observer memories are not a threat to IEM. In contrast, for another species of observer memories, the rememberer’s self is both part of the phenomenal content and intentional content. This species of observer memory poses a threat to IEM in a qualified way. In particular, whether this species of observer memory is a threat to IEM or not depends on the type of information that is used to construct the representation of the self in such images.

Zoom meeting room details:

Link:

https://ruhr-uni-bochum.zoom.us/j/97204546502?pwd=RmxsYUdRSzNzVVhiY3o3L0xpSnd3UT09

Meeting ID: 972 0454 6502

Password: 671485

In this article, I outline various ways in which artifacts are interwoven with autobiographical memory systems and conceptualize what this implies for the self. I first sketch the narrative approach to the self, arguing that who we are as persons is essentially our (unfolding) life story, which, in turn, determines our present beliefs and desires, but also directs our future goals and actions. I then argue that our autobiographical memory is partly anchored in our embodied interactions with an ecology of artifacts in our environment. Lifelogs, photos, videos, journals, diaries, souvenirs, jewelry, books, works of art, and many other meaningful objects trigger and sometimes constitute emotionally laden autobiographical memories. Autobiographical memory is thus distributed across embodied agents and various environmental structures. To defend this claim, I draw on and integrate distributed cognition theory and empirical research in human-technology interaction. Based on this, I conclude that the self is neither defined by psychological states realized by the brain nor by biological states realized by the organism, but should be seen as a distributed and relational construct.

People have vivid, confidently held memories of the circumstance in which they learned of a public, emotionally charged event, as well as memories about the event itself. The dynamics though which these memories are formed and retained are investigated in the context of the attack of September 11, 2001 in the United States. The results of a 10-year longitudinal study, a study on intergenerational transmission of memories, and a study on ageing will be discussed. The role flashbulb memories play in allowing people to give witness to public tragedies will be explored.

Our memories are not static. Each attempt to retrieve a past event can adaptively change the underlying memory space. Here I discuss my work on the neurocognitive mechanisms that enable the selective retrieval of episodic memories. I present behavioural and electrophysiological (M/EEG) work that provides insight into how a memory trace unfolds in time during retrieval, on a sub-trial scale. Further, I show evidence from a series of fMRI studies in which we track the representational changes that occur in a memory trace over time and across repeated retrievals. The latter findings demonstrate that retrieval adaptively modifies memories by strengthening behaviourally relevant and weakening behaviourally irrelevant, interfering components. Together, this work sheds light onto the neural dynamics of the retrieval process, and informs theories of adaptive memory.

In recent years, analysis of functional magnetic resonance imaging (fMRI) data in humans has expanded to include approaches focusing on the relational content in patterns of fMRI data rather than single-voxel activation profiles. Among these so-called multivariate pattern analysis (MVPA) techniques, representational similarity analysis (RSA) is the one that has most often been used to explore how semantic structure is represented in similarity patterns in the brain. Interestingly, RSA has also been used to show that when previously unrelated elements become linked through episodic memory, brain regions such as the hippocampus respond with changed pattern similarity for these elements. Thus, this analysis approach is well suited to investigate the interaction between semantic and episodic memory. In my talk, I will first give a general overview over what RSA is and how it has been applied in previous studies. Then, I will present ideas for the experiments in our project P4, and show some preliminary results from piloting studies.

The first retreat of the research unit FOR2812 will take place on Sept. 27th and 28th 2019. We will discuss preliminary results and plan the workshop that is to take place in 2020, as well as install the collaborations between the subprojects.