Mīmāṃsā authors think that cognitions are by themselves, i.e., intrinsically valid. In other words, they are not valid because of some additional reason, but just because of the sheer fact of being unfalsified.
In order to establish this claim, Veṅkaṭanātha defeats three possible candidates as extrinsic reasons for the cognitions’ validity, namely agreement (saṃvāda, with other people’s cognitions or with one’s other cognitions), quality of the cognition’s cause and causal efficacy (arthakriyā). The latter candidate is rejected because it would lead to an infinite regress (one would need to establish the causal efficacy of the cognition of causal efficacy and so on) and because some cognitions just don’t have any causal result. The passage is as follows:
arthakriyā ca na vyāptā, aśakyārthakriyeṣv anādarapadeṣu ca pratīteṣu puruṣapravṛttyabhāvena tanmūlārthakriyābhāvāt.
Nor is the causal efficiency (arthakriyā) included (vyāp-) [in the ascertainment of a cognition’s validity], because there is no causal efficiency as the root of certain [cognitions], given that there is no human activity (pravṛtti) in the case of apprehended [cognitions] in regard to which causal efficiency (arthakriyā) is impossible and whose contents (pada) have not been taken into account (anādara). (SM ad PMS 1.1.5)
UPDATE: The translation has been updated according to the suggestion (see comments below) by Lalitālālitaḥ (for which I am extremely grateful). I am sure he is right about the general meaning, but I would have still preferred anādarapadārtheṣu or anādaraviṣayeṣu.