The importance of interaction is increasingly emphasized in teaching in the new Finnish core curriculum for basic education. Working methods that activate students have become more common and inquiry-based learning has challenged the usual teacher-led teaching. The latest findings show that formative assessment has a positive effect on both teaching and learning. Formative assessment enables teaching to be focused and modified during the learning process. The aim of this study was to find out the presence and quality of discussion cycles in teacher-led and inquiry-based learning and to study the effects of distractions on formative assessment. Discussion cycles refer to ESRU cycles (Elicit, Student response, Regocnize, Use) which are discussion structures that are used during lessons. Previous studies have shown that they improve learning outcomes, and in this study they act as indicators of quality in formative assessment. Data of this study was gathered during five lessons of secondary school biology. The discussion cycles between the students and the teacher during the lessons were analysed. The study questions were 1) how does the teacher's previously given instruction for open questions affect the number and quality of the ESRU cycles? and 2) are there any distractions in inquiry-based learning that causes disruption of discussion cycles despite the fact that the teacher consciously tries to create them and find out the level of understanding of the students? The questions asked by the teacher during the lessons were categorized into open-ended and closed-ended questions. Instructions given to the teacher decreased the appearance of ESRU cycles. On average there were more disturbances in inquiry-based learning compared to teacher-led instruction. Prior to the given instructions the discussion cycles interrupted by disturbances were detected in both teaching methods. Instructions that were given to teacher in advance and the conscious creation of discussion cycles reduced the disruption of discussion cycles due to distractions. The qualitative nature of the study and the small sample size obviously partly affected that contrary to the hypothesis, the instructions for formative assessment, open questions, and the maintenance of discussion cycles reduced the number of discussion cycles during lessons.