Her choice will be from the pool of those males with a higher mate-value than that possessed by her pair-bonded partner, as is shown by women behaving in the very opposite way to men in their 'standards' re opposite-sex mate-value when looking for 'casual' sex [Szepsenwol, Mikulincer & Birnbaum 2013].
There is no contradiction between a universal motivation to pair-bond and the failure of some or many individuals ever to do so -- all too obviously, pair-bonding can be thwarted in so many ways that lifetime failure to form (or to sustain beyond the initial stages) a pair-bond is commonplace.
And it would be difficult to make an adaptive case for an unconditional open-ended affiliation.
Therefore, if the confrontation resolved to simple contest, most likely the male attempting to defend against cuckoldry would be defeated.
The physiology and psychology concerning dominance hierarchy provide evolved mechanisms to avoid initiating or escalating conflict with higher- (but not lower-) ranking males, thus likely avoiding from the outset the possibility of an escalation.
The popularly held notion that the pair-bond is naturally of very long duration is through conflating pair-bonding with other forms of affiliation that may augment and/or replace pair-bonding to change the nature of the relationship – a usual development to which the parties to long-enduring marriages often readily attest.
Ryan & Jetha do not acknowledge (even when repeatedly requested) [personal communications 2009, 2010] Fisher's cross-societal finding [Fisher 1989, 1994] that the facility to pair-bond indeed is universal, though with a short average duration of four years (based on the median age of separation after marriage).
With the consequences of sex for the female, there would be no value in extra-pair sex with any male of a lower mate-value than that of the pair-bonded partner.
If a male were to try to oppose the female partner's chosen extra-pair sex, then the cuckolding male, in being comparatively higher in mate-value, is likely to be the physically more powerful and/or more belligerent of the two, and/or to be a member of a stronger male coalition.
Manson [Manson 1997] reviewed the literature on pair-bonding across species and found that the courtship element of pair-bonding is predictive of reproductive outcome, but that this is not true of the element of mate-guarding.
A review of avian mating systems found little relationship between the strength of mate-guarding and paternity [Johnson & Burley 1998].
With pair-bonding primarily of benefit to the female, the requirement to mate-guard as hitherto understood, to prevent partner defection, is not performed by the male but by the female.