Shun Iizuka

The child objection is a major challenge for reductionism, which requires hearers to have positive reasons for testimonial justification. However, it has been pointed out that anti-reductionism, which requires only the absence of negative reasons, or defeaters, suffers from the same kind of problem. The child objection presupposes the empirical thesis that “children do not have the capacity to consider reasons,” but the plausibility of this assumption may be revealed by developmental psychology research on selective trust. This paper uses recent epistemological studies as a guide to narrow down the types of defeaters that children are required to consider, and then clarifies what kind of reasons various experiments can be said to test the ability to consider, and in what sense children who pass the test can be said to be “considering” reasons. In doing so, we clarify the scope and limits of the implications that selective trust studies can have for reductionism and anti-reductionism. We then suggest what future psychological research is desired from an epistemological interest to go beyond the current limitations.

Felix Lo offers suggestions for the further development of the ideas in my paper, “Taking It Not at Face Value: A New Taxonomy for the Beliefs Acquired from Conversational AIs” (2024) which appeared in Techné, through several critiques. Lo’s reply, “Interrogating the Testimony of AIs” (2025), consists of two main parts. The first half provides a thorough and accurate summary of my paper. In the latter half, Lo points out several oversimplifications and misunderstandings in my argument and, in doing so, expresses concern about expecting our “filtering abilities” to be good enough. I agree with the individual points raised and believe that readers will gain a more accurate understanding of the technology and the related concepts by reading Lo’s reply in addition to my paper. I am also convinced that these points will affect the value of my proposal. Therefore, rather than refuting Lo’s points, what I should do in this paper is to organize how, to what extent, and to which part of my argument they apply. In doing so, I will clarify the direction of further research.

One of the central questions in the epistemology of conversational AIs is how to classify the beliefs acquired from them. Two promising candidates are instrument-based and testimony-based beliefs. However, the category of instrument-based beliefs faces an intrinsic problem, and a challenge arises in its application. On the other hand, relying solely on the category of testimony-based beliefs does not encompass the totality of our practice of using conversational AIs. To address these limitations, I propose a novel classification of beliefs that shifts the focus from the properties of the conversational AIs themselves to how recipients perceive AI output, specifically whether they take the output at face value. The proposed categories are beliefs by regarding as instruments and beliefs by regarding as testifiers. By using these complementary categories, a more comprehensive understanding of beliefs acquired from conversational AIs can be achieved while avoiding the problems associated with the traditional classification.

The recursive aspect of process reliabilism has rarely been examined. The regress puzzle, which illustrates infinite regress arising from the combination of the recursive structure and the no-defeater condition incorporated into it, is a valuable exception. However, this puzzle can be dealt with in the framework of process reliabilism by reconsidering the relationship between the recursion and the no-defeater condition based on the distinction between prima facie and ultima facie justification. Thus, the regress puzzle is not a basis for abandoning process reliabilism. A genuinely intractable problem for recursive reliabilism lies in the gap between the reliability of the entire path to a belief and that of its parts. Confronted with this puzzle, reliabilists can orient themselves toward ‘reliable-as-a-whole reliabilism’ instead of ‘reliable-in-every-part reliabilism’, including recursive reliabilism, which is found to be not well-motivated.