THE UNDO EFFECT OF POSITIVE EMOTIONS

Abstract
Negative emotions can be detrimental to certain aspects of memory, an effect not seen with
positive emotions. Elicitation of positive emotions following a negative emotional experience
may “undo” detrimental effects of negative emotion on memory. The current study examined
whether recall of a narrative describing a traumatic experience would be better if it had a positive
ending versus a negative ending. Participants read an account of a drowning (i.e., traumatic
experience) and subsequent life-saving attempt that ended with either death (negative emotion
ending) or resuscitation (positive emotion ending). On recall, the positive ending group
demonstrated better memory than the negative ending group (p < .05). Elicitation of positive
emotion after a traumatic experience may reduce deleterious effects on memory and may have
implications for the treatment of post-traumatic stress disorder. 

Past research on emotional memory has paid more attention to traumatic and negative
emotional experiences with little attention to positive ones. There is a general consensus among
researchers that memories for trauma are inherently different from other types of memories (see
Brewin, 2007 for a review; Ferree & Cahill, 2009; Paz-Alonso & Goodman, 2008; Wegner,
Quillian, & Houston, 1996 ). In the literature, memories for negative and traumatic emotional
experiences suffer from chronological disorganization and fragmentation (Brewin, 2007; Ferree
& Cahill, 2009; Kindt & van den Hout, 2003; Wegner et al., 1996). Kindt and van den Hout
report that individuals who have experienced trauma have difficulty organizing the memory into
a coherent narrative with an ordered sequence of events. It has been suggested that the high level
of emotionality that accompanies a traumatic experience is to blame for the memory disruptions
associated with the experience of trauma (Ferree & Cahill, 2009; Wegner et al., 1996).

Bohanek et al. (2004) rated written narratives of recalled autobiographical memories of
positive and negative emotional events on a 4-point scale with regard to organization, clarity,
amount of detail, chronology, completeness, and understandability. They found that negative
recalled memories contained more negative emotion words, cognitive processing words, and
passive sentences than positive memories. Positive recalled memories, on the other hand,
contained more positive emotion words and were more complex than negative memories.
Furthermore, participants rated negative events as more emotional than positive events. Intensely
positive memory narratives were the most coherent, while intensely negative memory narratives
were lacking in coherence and complexity, although they were longer. Overall, memory for
trauma was found to be less coherent than positive memories and suffered from chronological
disorganization. 
Berntsen (2002) found that narratives of recalled negative memories contained more
central details than peripheral details and less peripheral details than narratives of positive
memories. She offers an evolutionary perspective for this difference, stating that it makes sense
that one would concentrate more on central details than peripheral ones when experiencing fear
than when experiencing happiness. Fear may cause a tunnelling effect on memory (Berntsen,
2002; Safer, Christianson, Autry, & Osterlund, 1998) in which the mind focuses only on those
details that are important for survival. Attention and memory narrow in on central details, at the
expense of peripheral details, which are lost to conscious awareness. Berntsen (2002) was unable
to find any evidence that tunnelling occurs in positive emotional memories, instead positive
emotional memories contained slightly more peripheral details than central details. This upholds
the notion that traumatic and negative emotional memories are fundamentally different from
positive memories.

In the past decade, positive emotions have gained notice in the research community and
are being studied for their health benefits. Positive emotions promote health and longevity and
broaden attention, cognition, and behavioural ranges (Fredrickson & Cohn, 2008). Fredrickson
and Cohn (2008) assert that better coping is observed in individuals who have experienced
trauma if some level of positivity accompanies their thoughts and feelings regarding the trauma.
Research in the areas of motivation, aggression, and anxiety disorders suggest that key aspects of 
THE UNDO EFFECT OF POSITVE EMOTIONS 6
positive and negative emotions may be incompatible, although the mechanism responsible for
this has not yet been defined (Fredrickson & Cohn, 2008). 
Fredrickson and Cohn (2008) propose that positive emotions have an "undo effect" on the
experience of negative emotion. Fredrickson, Mancuso, Branigan, and Tugade (2000) induced
high arousal negative emotion followed by either positive emotion, negative emotion, or no
emotion (neutral). A faster rate of cardiovascular recovery was observed in participants in the
positive emotion condition than in the neutral condition and those in the neutral condition
recovered faster than those in the negative condition. Furthermore, when the positive emotion
film clips were observed without prior induction to a negative state, no cardiovascular effects
occurred at all (Fredrickson et al., 2000). Thus, Fredrickson et al. (2000) suggested an undo
effect of positive emotion on the cardiovascular effects of high arousal negative emotion. In
addition to this, self-reported levels of psychological resilience have been correlated with rate of
cardiovascular recovery (Block & Kremen, 1996). It is conceivable, then, that the undo effect
may be seen in memory for highly emotional negative or traumatic experiences.
Few experiments have been performed with traumatic representations in literature, rather
than film. In fact, research surrounding reading and narrative processing is rather neglected
overall (Oatley, 1999). Evidence has emerged that reading fiction involves similar processes of
cognition and evokes similar emotions as one would encounter in real life (Mar, Oatley, Djikic,
& Mullin, 2011; Oatley, 1999). Furthermore, although analogue trauma films may fulfil DSMIV-TR requirements for being witness to a trauma, literary depictions of trauma, by way of
narrative engagement and imagination, can place a reader in the midst of the situation within her
own mind. Narrative engagement, also referred to as narrative transportation (Gerrig & Rapp,
2004), is the mechanism by which the reader engages in the narrative as if it were an actual
event, simulating the events of the narrative in the mind (Mar, Oatley, Hirsh, dela Paz, &
Peterson, 2006).
Research indicates that reading about an experience activates the same cognitive
processes and emotional responses as an actual real-life experience (Boulanger, Hauk, &
Pulvermüller, 2009; Gerrig & Rapp, 2004; Mar et al., 2006; Mar, Oatley, & Peterson, 2009; Mar
et al., 2011; Oatley, 1999; Zwaan & Taylor, 2006). Oatley (1999) proposes that fictional
narratives are simulations that run on minds. 
References
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorderstext revision, DSM-IV-TR, (4th ed.). Arlington VA: American Psychiatric Publishing, Inc.
Baron-Cohen, S. (2003). The essential difference: The truth about the male and female brain.
New York, NY: Basic Books.
Baron? Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, I. (2001). The “Reading the
mind in the eyes” test revised version: A study with normal adults, and adults with asperger
syndrome or high? functioning autism. Journal of Child Psychology and Psychiatry,
42(2), 241-251. doi:10.1111/1469-7610.00715
Berntsen, D. (2002). Tunnel memories for autobiographical events: Central details are
remembered more frequently from shocking than from happy experiences. Memory &
Cognition, 30(7), 1010-1020. doi:10.3758/BF03194319