Do you remember who you are? (Nada Gligorov)

Propranolol, an inexpensive and widely available beta blocker, is known to be effective in preventing post traumatic stress disorder (PTSD) if administered soon after a traumatic experience. In a recent article, Elsey and Kindt (2016) describe a new way of using propranolol to diminish PTSD symptoms in individuals who already have the condition. The authors describe research that indicates that propranolol can be effective in treating PTSD when administered during the reconsolidation of long-term memories. Taking propranolol as the memory is retrieved diminishes the emotional bite of traumatic memories without erasing them, and without significantly affecting declarative memory for the event.

When memory is characterized as foundational to narrative identity, memory modification has been seen as having morally perilous consequences because of its potential to change or even disrupt identity. But there is reason to doubt that memory modification can have this effect.  In a series of studies, Loftus (2003) has shown that memory is reconstructed at the time of recall, and that it is vulnerable to retroactive interference.  A memory can be altered in a number of different ways: it can be influenced by how we are questioned about the witnessed event; it can be contaminated by false information presented after the happening; and it can be changed through conversation about what was observed. Additionally, even autobiographical memories can be unreliable.  Sheen et al. (2001, 2006) showed that siblings who concur on the details of childhood memories will disagree about their ownership. If the memory was about achievement, the siblings were more likely to remember it as happening to them, and if it was about wrongdoing they were more likely to attribute it to their brother or sister.  This illustrates a further point, which is that instead of memory grounding our sense of self, it is actually the other way around--we have a narrative about who we are and we use it to reshape our past (Wilson and Ross 2003).  

Despite this evidence for the unreliability of witnessed and autobiographical memory, one could still maintain that an individual might require certain core memories to maintain a sense of continuity of their identity over time. However, a case of memory loss presented by Klein and Nichols (2012) indicates otherwise. This is a case of a man who suffered significant head trauma in a motorcycle accident and experienced retrograde and anterograde amnesia for events that occurred soon before and after his accident. This individual had an intact episodic memory for earlier events from his past, but he described remembering those events as if they had happened to somebody else. Nonetheless, he reported a continuity of his sense of self. Although the specifics of this case are unusual, they suggest that narrative identity does not depend on the maintenance even of core memories.

In their article, Elsey and Kindt (2016) are careful to argue only for the permissibility of the use of propranolol to treat diagnosed medical or psychiatric conditions. But if it is recognized that such memory modification may not be a challenge to narrative identity, there is room for a stronger conclusion. If effective and safe methods are available, memory modification is morally permissible even beyond the treatment of disease, but more broadly when used toward improvement in quality of life.

Comments

[Adam Kolber]

Thanks, Nada! I didn't know about the Klein and Nichols paper. Very interesting. I'm inclined to agree with your conclusion in the last paragraph, though I think there are several other potential concerns aside from narrative identity: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=887061

(You don't explicitly say that narrative identity is the only concern but perhaps you suggest it by saying that if concern over narrative identity is surmounted we can reach the stronger conclusion you make. Also, I guess it may depend on how broadly we construe the term narrative identity in terms of deciding which potential ethical concerns can be attributed to it.) Again, a very helpful post and I like the Elsey and Kindt paper, too!

[Nada Gligorov]

Thanks, Adam.
I do not think that narrative identity is the only concern, but it is a relatively unique concern raised by memory modifiers. I focused on the impact of memory modification on narrative identity because it is occasionally used to argue against the general permissibility of memory modification. And I think that based on the evidence presented in the post, there is no reason to assume that memory modification would result in a disrupted narrative identity. Of course individual prescription decisions have to be made taking into account the risks and benefits to individual patients, but propranolol is a relatively safe medication. With regard to my stronger conclusion, I think it follows from the argument about the general permissibility of memory modification because I think that the treatment and enhancement distinction cannot be drawn. Once one accepts the general permissibility of propranolol for the treatment of PTSD, then I think there are no general moral barriers for its use to improve quality of life for those who are not diagnosed with that disorder.

[Adam Kolber]

Sounds good to me! (And I agree with your view about the therapy/enhancement distinction.)

[Andrea Lavazza]

As I pointed out in my commentary in “Ajob Neuroscience” (Lavazza, 2016 http://www.tandfonline.com/doi/full/10.1080/21507740.2016.1251988), manipulating human memory through reconsolidation is an important new frontier of research in the therapeutic field. The ethical implications of these specific techniques to treat traumatic memories have been appropriately analyzed by James and Kindt (2016 http://www.tandfonline.com/doi/full/10.1080/21507740.2016.1218377). They say that hypothetical moral scenarios should be realistic and speculations about the future cannot ignore the data and the restrictions provided by the research. However, there do seem to be real ethical concerns in relation to non-clinical memory manipulation. I illustrated three of them that are worthy of consideration.
The first (Erler, 2011 http://link.springer.com/article/10.1007%2Fs12152-010-9090-4) concerns interventions that threaten the authenticity of the subject’s choices (if one "fades" the unpleasant memory of wrongs received, one can forgive one’s molesters without asking for justice, which one would have done without the memory manipulation). The second (Glannon, 2011) refers to situations involving the risk of losing touch with reality (a young researcher who mitigates the memories of his faux pas to be able to move on sooner or later will come up against total failure). The third (Lavazza, 2015 http://peh-med.biomedcentral.com/articles/10.1186/s13010-014-0021-6) has to do with social composition effects (if an abused woman legitimately resorts to memory manipulation to alleviate her suffering, when the majority of women in her condition does the same the motivational drive to rebellion and social change will weaken, as it is rooted in emotionally painful memories).
None of them makes reference to “narrow” narrative identity, but each of them considers a “wide” or “public” narrative identity which is also linked to relevant pieces of autobiographical memory and, in my opinion, cannot be manipulated or disrupted without paying a price.
Andrea Lavazza

[Nada Gligorov]

Thank you for your comment, Andrea.
I actually address some of the examples you consider in chapter 5 of my recent book (http://www.springer.com/us/book/9789402409642). I disagree with Erler (2011) and argue for a subjectivist conception of authenticity.
But more to your point about the impact of memory modification on social obligations: I think that those types of ethical considerations should be made by the individual considering the use of propranolol. Given that propranolol does not affect declarative memory, it is not clear that social activism against abuse would be affected by taking it. An abused individual would still remember having been abused, would still remember having to deal with the consequences of that event, and then also remember having to use propranolol to alleviate suffering. All that could still result in a commitment to social activism against abuse. After all, people who’ve never been abused are committed to fighting abuse. But more generally, it strikes me that each person considering memory modification ought to be able to decide whether the risks (and here risks would be construed broadly to include some of the social consequences you mentioned) outweigh the benefits of being relieved of suffering. My argument was for the general permissibility of the use of propranolol, which is compatible with there being instances in which it is immoral for an individual to modify their memories. The example of abuse, however, doesn’t strike me as such an example, especially if we take into account that types of abuse can actually cause PTSD.