Table 2. Guidelines for infant fNIRS reporting.

	Strongly recommmended	Additional
Participants	Broad infant population type, state (involving sleep or awake), number included and excluded, criteria for exclusion, mean and range of chronological age, sex composition, language background in case of language study	Hair composition, skin pigmentation composition, (handedness composition of parents, for studies on lateralization), mean and range of gestational age at birth, whether sleep state is quiet sleep or active sleep in the case of sleeping infants
Instrumentation	Type of NIRS (e.g. CW system, time-resolved spectroscopy), Brand and model, power (mW per physical source), number of sources, number of detectors, number of channels defined, interoptode separation(s), sampling rate	Pad geometry, pad localization relative to an anchor in 10–20 space, mean and range of any head measurements that have been taken
Stimuli and procedure	General type of stimulation, general characteristics of baseline, mean and range of stimulation duration, mean and range of preceding and following rest duration, mean and range of number of 'good' stimulation blocks an infant*channel must have to be included	Total duration of the study, hyperlink to where (a sample of) the stimuli are stored
Pre-processing	Specific details on data processing for pre-processing, artifact detection, and removal, as detailed below. In all cases, specify to which signal they were applied; and whether they were applied by channel or channel group (e.g., optode- or pad-based). If relevant, analysis package used (e.g. HOMeR, POTATo)	Hyperlink to the scripts used throughout the pre-processing pipeline, if possible POTATo ‘recipe’ [12]
	1. Pre-processing and detrending: if filtering used, frequency and type of filter for low and band-pass; if baseline level fit, length of the preceding and following baseline, and type of fitting; in GLM, periods of sines/cosines declared for detrending
	2. Artifact detection: for methods based on rapid changes, whether identification was automatic or manual, window size in seconds and criterion change; for PCA-based methods, method for selecting the principal component
	3. Artifact removal: whether the whole block is excluded, only the artifacted stretch is removed, or whether the artifacted stretch is replaced (if so, what specific type of interpolation)
Activation analyses and results	Report both oxyHb and deoxyHb; if using an ROI, explain how it was selected; if restricting the analyses to a temporal window, explain how it was selected; state which method was used to derive the dependent measure: GLM, average, peak (standard or absolute?), area under the curve; note if any further processing (e.g., z-scoring) was done to the data. Include a figure with the average time course and SE bars of oxyHb and deoxyHb changes for each channel	Hyperlink to the scripts used throughout the analysis pipeline, method of spatial estimation of brain region (e.g. photogrametry, 3D digitizer), hyperlink to where the original data are stored in a common format (e.g., if possible.snirf, [13])
Additional standardized signal-to-noise estimation	Using denoised data (but without z-scoring or converting it using absolute first), calculate, for each individual infant, oxy-Hb and deoxy-Hb peak amplitude within a time window starting at the onset of stimulation and ending at the shortest interstimulus interval used; within all channels in a ROI which plausibly includes a brain region that responds to the kind of stimulation used. Extract also the time of this peak response. Calculate the standard deviation across infants for this peak response for each individual channel. Then determine which of these channels has the greatest signal strength defined as the average peak amplitude divided by the standard deviation across infants for oxyHb and deoxyHb – this will be the selected channel. If the greatest signal strength according to oxyHb and deoxyHb is found in different channels, select both. For the selected channel(s) report: precise interoptode separation, localization relative to an anchor in 10–20 space, mean peak oxyHb, SD peak oxyHb, mean and SD latency of this peak oxyHb, mean peak deoxyHb, SD peak deoxyHb, mean and SD latency of this peak deoxyHb

Items in 'strongly recommended' are necessary to achieve basic standardization and thus we should strive to report them; those in 'additional' would also be included for completeness.