Serendip is an independent site partnering with faculty at multiple colleges and universities around the world. Happy exploring!

Don't Cross Mama Bear: A Central Pattern Generator in Maternal Aggression

EB Ver Hoeve's picture

Equipped with the tool of the central pattern generator, we can explain how the architecture and actions of the nervous system translate into behavior. “Stored motor symphony scores” explain the existence of most patterns of action, and most of the observations discussed in class and on the forum can be grouped within one of two categories. First, observations of characteristics, such as walking in vertebrates, suggest an innate genetic influence. Second, observations of activities, such as masterfully playing the piano, suggest acquisition of central pattern generators from muscle memory based on repeated learned movements. But could there exist a third category of behaviors that can be explained through the existence of a central pattern generator, but that does not belong exclusively to either of the offered possibilities for how a central pattern generator can be built out of neurons?

The CPG for female aggression in mice, which does not become observable until after pregnancy and during lactation, may provide an example. Unlike flying or any other inborn complex output involving a genetically required CPG, maternal aggression in lactating mice may provide potential cause for investigation of how central pattern generators are created. Perhaps the CPG for maternal aggression can be explained as a combination of our previous two theories: a modulated interaction between the hormonal levels (environmental input) and gene expression (dormant genetic influence) which after pregnancy, patterns maternal aggression to be reflected as a stored motor symphony.

In this paper I bring my new understanding about central pattern generators to my previous experience studying maternal aggressive behavior in mice in Dr. Stephen Gammie’s neurobiology lab (UW-Madison). Much like our class’ decision to initiate investigation of behavior by examining outputs and then working backward toward inputs, my first job in the lab was to evaluate outputs of aggression in female mice. Aggression tests involved direct removal of mouse pups from the cage of a lactating female followed by immediate introduction of an intruder male mouse. For 10 minutes, the female’s behaviors were systematically charted. Inevitably, within seconds after the male mouse’s feet touched the ground, the female attacked. Often springing from behind and sinking her teeth repeatedly into his backside, the female mouse appeared prepared to fight to the death. Although difficult to articulate – especially when I was performing the test – there came a point when I could predict when and where the female would attack. After repeatedly watching this pattern of aggressive responses in a series of lactating female mice, it became clear that in order to protect their offspring, these female mice unconsciously leapt into a patterned aggressive response.

Interestingly, although lactating mice expressed high aggression toward the male in this context, virgin mice displayed little or no aggression. Upon introduction of the male mouse, virgin female mice backed into a corner and focused on how to stay out of the intruder’s way. Qualitatively, virgin females appeared fearful and anxious, behaviors that can also be categorized and predicted, thereby suggesting the result of a central pattern generator.

Aggression represents a complex behavioral phenomenon consisting of both mental and physical components that can neither be reduced to a linear process nor evaluated from a single angle. When I discuss findings from the research I conducted showing a connection between becoming a mother and exhibiting increased aggression, people generally respond knowingly saying, “Yes, it is wise not to stand between a mother bear and her cubs!” Only now, when I reflect on my research observations within the context of my new understanding about the architecture and tools of the nervous system am I able to see that the creation of the central pattern for maternal aggression is likely a combination between genetics and the environment. In fact, these findings suggest that pregnancy and birth may be responsible for triggering biological changes in gene expression and neurotransmission in the brain, thereby leading to a central pattern of increased maternal aggression.

If perceptions are conceived in the brain, and the brain has been observed to undergo physiological changes following pregnancy and birth, then it makes sense that the female mouse who previously perceived the intruder male as frightening, could later perceive him as a threat. From these contrasting observations we are left with several questions: If maternal aggressive behavior represents a stored motor symphony score, how was this central pattern generated? If the explanation was as simple as a genetic influence, why/how did the central pattern for behavioral aggression switch?

For the lactating female mouse, increased aggression is just one of the behavioral trait changes that occur during pregnancy and birth. Changes in the ability of the female to produce milk and to nurse her young are also linked to physiological changes in the brain that lead to changes in behavior during lactation. Scientists can attribute some changes to hormonal influences, but the endogenous changes in neurotransmission associated with aggression are still being explored. One of the receptors thought to be involved is GABA. Known as a primary inhibitory neurotransmitter in the CNS, increasing GABA is pharmacologically understood to lead to reduced anxiety. The research lab where I worked explored this connection between GABA enhancement and increased maternal aggression using the GABA agonist, chlordiazepoxide (CDP). CDP has been shown to significantly increase maternal aggression (Lee and Gammie, Pharmacology Biochemical Behavior 2007). Studies of changes in brain activity have suggested that CDP may be necessary (though probably not sufficient) in the observed behavioral change. That is, biological changes within the brain cause an increased activation or inhibition of specific genes controlling maternal aggression. These biological changes within the brain are initiated by the body’s response to pregnancy and therefore must partially be understood as the body’s unconscious input in creating the CPG for maternal aggression.

Extensive gene expression studies further suggest that there may be a genetic component to this switch in aggression controlled by a developmental shift that occurs during pregnancy. Believed to be an important mediator in the stress response, corticotropin-releasing factor is being studied as an indicator of changes in behavior. Without CRFR1 (knockout mice), maternal behavior – especially nurturing – decreases. This suggests that CRFR1 is required for complete maternal behavior, and yet is not the only gene involved in the creation of this central pattern (Gammie et al, BioMed Central 2007). Gene expression studies demonstrate that several genes become more expressed during lactation, which in turn implies that the creation of the CPG for maternal aggression must partially be based on genetic influence.

This paper is not intended to provide an answer for how the CPG for maternal aggression is created. Even if studies provided an explanation for the biological changes that the brain undergoes during pregnancy, the CPG for maternal aggression would still not be fully understood. Instead, this paper attempts to point out observations that suggest there may exist alternatives for how central pattern generators are built. Furthermore, the observations imply that this behavioral shift is not conscious. While the acquisition of CPGs typically involves a conscious decision to initiate a specific action, maternal aggression is an unconscious behavioral shift. In fact, thinking about aggression as an unconscious behavioral shift may hold significance for our society. If it is possible to have unconscious behavioral shifts, then individuals who exhibit aggression should not necessarily be blamed. Without input, the brain is capable of generating output. So in the end, the notion of “doing something without knowing why you did it” is perhaps not as crazy as it once sounded.





Lee, Grace. Gammie, Stephen. “GABA Enhancement of Maternal Defense in Mice: Possible Neural Correlates” Pharmacology Biochemical Behavior. 2007 January; 86(1): 176-187. Pubmed. 4/4/08

Gammie, Stephen. Bethea, Emily. Stevenson, Sharon. “Altered maternal profiles in corticotropin-releasing factor receptor I deficient mice” BioMed Central. March 2007, 8;17 1471-2202/8/17


Paul Grobstein's picture

female aggression and life

I suspect it is indeed true that there is an "unconscious behavioral shift" here, one involving hormones, but how could one know in a rat whether it is "unconscious" or not? And if unconscious in humans does that necessarily mean "individuals who exhibit aggression should not be blamed"?