You are probably referring to Haselton & Buss's Error Management Theory (sometimes referred to as the 'smoke detector principle'; e.g. by Randy Nesse)

Hello, not sure what you mean with "prioritize" negative memories. Healthy people have more frequent positive memories. Individuals suffering from PTSD report higher frequency and/or more distressing intrusive memories related to a specific traumatic event, but this is of course different from the normal population. Individuals suffering from depression also have a bias towards recalling negative info. This again, not the norm.

With "prioritizing" you might mean that single highly emotional events (ie., very negative, traumatic) are easily recalled later on or leave a big "mindprint". However, this is not *only* for negative events, it is also for highly positive events. Having easy access to memories of negative unique events might serve an evolutionary purpose (ie., preventig, protecting), but it does not mean that our overall cognitions are predominantly negative. It is much the opposite, positive cognitions/memories dominate psychologically healthy minds. Many good scientific studies here

http://conamore.au.dk/en/publications/

In prehistoric times, catching a rabbit for dinner was positive and being eaten by a sabertooth tiger was negative. If you missed a positive experience you were uncomfortable all night but survived. A negative experience (being eaten) prevented all future positive experiences. So our minds are programmed to avoid negative more than to seek positive. Hence, if you say something negative to your spouse, it must be negated with four or five positive words because the negative word is stronger.

Except that is an oversimplified view of what and how people remember. I think what you are trying to say is that memories of certain events have greater psychological/physiological impact than others, which is true. However, Valence (positive vs negative), is only one dimension of events and their memories. Arousal, recency, centrality of event for one's identity, and current mood, are all factors that determine what memories are more easily accessed. Being at risk of being eaten by a sabertooth is a life-threatening situation. Thus differing markedly from other negative, non traumatic experiences. In our times the situation would be analogue to be put a gun in front of your face during a robbery. Thus, this event is unique, highly stressful/arousing , not "simply" negative. And it activated a fear response that becomes conditioned. It is different from other negative events (ie., finding that the ATM is broken) in a number of dimensions. I think what you are trying to say is that memories of certain events have greater impact than others, and this is only partly determined by their valance. Negative events have greater likelihood to provoke higher arousal, but that does not mean we are bombarded by  memories of these events later on (only you suffer from a psych disorder such as PTSD). Memories of positive events are easily remembered in healthy populations.  http://onlinelibrary.wiley.com/doi/10.1002/acp.838  Highly emotional memories might be easily accessed when the memory becomes relevant for your current context, regardless of their valence.   The 1:5 ratio is an incomplete explanation of real cognitive activity. It is like the magic number 7  + - 2 for chuncking....It is true, but only a partial truth

Vivid recollection of negative information has been widely documented (Dewhurst

& Parry, 2000; Rimmele, Davachi, Petrov, Dougal, & Phelps, 2011; Sharot & Yonelinas, 2008). Yet as Adriana pointed out, valence is defined as the level of pleasantness generated when a stimulus is encoded and varies from strongly negative to strongly positive. Memory for items with EITHER negative or positive valence is usually better than for neutral items (Kensinger & Corkin, 2003; Ochsner, 2000). 

Dewhurst, S. A., & Parry, L. A. (2000). Emotionality, distinctiveness, and recollective experience. European Journal of Cognitive Psychology, 12, 541–551.

Kensinger, E. A., & Corkin, S. (2004). Two routes to emotional memory: Distinct neural processes for valence and arousal. Proceedings of the National Academy of Sciences of the USA, 101, 3310–3315.

Rimmele, U., Davachi, L., Petrov, R., Dougal, S., & Phelps, E. A. (2011). Emotion enhances the subjective feeling of remembering, despite lower accuracy for contextual details. Emotion, 11, 553–562.

Sharot, T., & Yonelinas, A. P. (2008). Differential time-dependent effects of emotion on recollective experience and memory for contextual information. Cognition, 106, 538–547.

Ochsner, K. N. (2000). Are affective events richly recollected or simply familiar? The experience and process of recognizing feelings past. Journal of Experimental Psychology: General, 129, 242–261.

You speak of the Prospect Theory by Kahneman and Tversky, (1979, 1992).

The fact that people prefer to avoid loses than achieve gain.

Hello,

Perhaps the "Bad is Stronger Than Good" paper by Baumeister et al. 2001 might help?

"Negativity bias" may be the term you are trying to remember.

I reckon the  terms could be '' less is more ''