Reconciling one-dimensional valence with distinct mechanisms for reward/punishment

Question

It is a widespread notion in psychology that there are two dimensions to emotion: valence (aversiveness or attractiveness of some entity) and intensity. However, recent experimental studies in neuropsychology (e.g. those listed below) have given strong evidence for two distinct pathways which process reward and punishment associated with some input. This in turn compellingly suggests that valence is in fact made up of two independent dimensions, and I'm guessing there are behavioral studies reaching similar conclusions (i.e. that persons are able to have "mixed feelings"). How do we reconcile these views? Are we forced to reject one-dimensional valence on the basis of the evidence?

Distinct pathways for reward/punishment:

• Paton, Joseph J., and Kenway Louie. "Reward and punishment illuminated." Nature neuroscience 15.6 (2012): 807. (http://www.researchgate.net/publication/225043743_Reward_and_punishment_illuminated/file/e0b495227b6d325c3d.pdf)

• Fujiwara, Juri, et al. "Segregated and integrated coding of reward and punishment in the cingulate cortex." Journal of Neurophysiology 101.6 (2009): 3284-3293.

• Monosov, Ilya E., and Okihide Hikosaka. "Regionally distinct processing of rewards and punishments by the primate ventromedial prefrontal cortex." The Journal of Neuroscience 32.30 (2012): 10318-10330.

Answer 1

A meta-analysis of 397 neuroimaging studies was just published looking at this very question (see here).

They tested three hypotheses:

1. Bipolarity - negative and positive are on the same continuum
2. Bivalence - negative and positive are independent
3. Affective workspace - valence is supported by a "flexible set of valence-general regions".

Their analyses are rather complicated, but they didn't find much support for the bipolarity or bivalence hypotheses. Instead, in support of the affective workspace hypothesis, they conclude (emphasis mine):

These findings suggest that, at the level of regional brain activity, there is no single region or even voxel that uniquely represents positivity or negativity. Limbic tissue, including the anterior insula, rostral ACC/ventromedial prefrontal cortex, dorsal ACC, amygdala, ventral striatum, as well as several other regions including the thalamus and occipitotemporal cortex, appears to contain cells that are part of the brain’s valence-general affective workspace or “affective neural reference space

But they note that some brain regions have preferences for negativity or positivity, meaning that these brain regions may tend to represent negativity or positivity over the other.

This meta-analysis isn't definitive, but it's the best we have right now.

Answer 2

I think that we are. But the notion of 'valence' is already problematic, as a variation on mathematical negation, which produces bizarre mathematical artifacts like Russel's Paradox.

It is tempting to follow natural language and find two balancing processes always add up to a sign and a magnitude. But in fact no two processes are ever going to consistently align in perfectly opposite directions, and in the gaps there is still interesting behavior. Something as clean as real negation only ever occurs in language and in theory.

We know from earlier studies of memory that positive or negative memories are the result of separate reactions from the back-brain and the frontal lobe, and that people can actually become attached to negative stimuli like physical pain for the pleasure they produce.

One would expect the same sort of balancing act, if not exactly the same process in a less direct form, to explain the effects of reward and punishment. So there will be places, parallel to the situation of attraction to pain, where reward and punishment do not align as opposites. (Situations of getting attention through punishment, or in its aftermath come to mind.) In such situations, the notion of valence only gets in the way.