Nutrition in autism spectrum disorders: A review of evidences for an emerging central role in aetiology, expression, and management

Olakunle James Onaolapo; Adejoke Yetunde Onaolapo; Olakunle James Onaolapo; Adejoke Yetunde Onaolapo

doi:10.3934/medsci.2018.2.122

AIMS Medical Science

2018, Volume 5, Issue 2: 122-144. doi: 10.3934/medsci.2018.2.122

Previous Article Next Article

Review

Nutrition in autism spectrum disorders: A review of evidences for an emerging central role in aetiology, expression, and management

Olakunle James Onaolapo ¹,
Adejoke Yetunde Onaolapo ^{2
,
,}

1.
Department of Pharmacology, Ladoke Akintola University of Technology, Osogbo, Osun State, Nigeria
2.
Department of Anatomy, Ladoke Akintola University of Technology, Ogbomoso, Oyo State, Nigeria

Received: 27 December 2017 Accepted: 09 March 2018 Published: 20 March 2018

Autism spectrum disorders (ASDs) are a group of neurodevelopmental disorders whose aetiology remains largely unknown, but for which environmental factors appear to be important. Emerging evidences suggest that nutrition may play a major role in the aetiology of ASD; also, specific maternal nutritional-deficiencies appear to be associated with an increased risk in offsprings. In addition, studies are beginning to reveal the beneficial effects of dietary supplementation or restriction in the management of ASD; while at the same time debunking the myths that surround certain purportedly-therapeutic dietary manipulations. In this narrative review (using information from internet databases such as Google scholar, PubMed, Scopus and authoritative texts), we examine the emerging central role of nutrition in relation to aetiology, symptomatology, management, and indices of outcome in ASD; by highlighting available scientific evidences pertaining to the impacts of different dietary manipulations and nutritional supplementation. We also consider the likely future roles of nutrition in ASD, as science continues to grapple with the understanding of a group of neurodevelopmental disorders that are emerging to be largely “nutritional illnesses”.

Keywords:

Citation: Olakunle James Onaolapo, Adejoke Yetunde Onaolapo. Nutrition in autism spectrum disorders: A review of evidences for an emerging central role in aetiology, expression, and management[J]. AIMS Medical Science, 2018, 5(2): 122-144. doi: 10.3934/medsci.2018.2.122

Related Papers:

[1]	Kevin Rebecchi . Beyond “autism spectrum disorder”: toward a redefinition of the conceptual foundations of autism. AIMS Medical Science, 2025, 12(2): 193-209. doi: 10.3934/medsci.2025012
[2]	Joseph L. Amaral, Veronica Romero, Heidi Kloos, Michael J. Richardson . Complexity and Autism Spectrum Disorder: Exploring Hysteresis in A Grasping Task. AIMS Medical Science, 2017, 4(1): 113-130. doi: 10.3934/medsci.2017.1.113
[3]	Douglas Wilson, Griffiths Keith, Buttar Harpal, Singh Ram, De Meester Fabien, Wilczynska Agnieszka, Takahashi Toru . Therapy through Social Medicine: Cultivating Connections and Inspiring Solutions for Healthy Living. AIMS Medical Science, 2017, 4(2): 131-150. doi: 10.3934/medsci.2017.2.131
[4]	Sylvia Kirchengast, Beda Hartmann . Maternal prepregnancy nutritional status influences newborn size and mode of delivery. AIMS Medical Science, 2018, 5(1): 53-66. doi: 10.3934/medsci.2018.1.53
[5]	Snigdha Misra, Yang Wai Yew, Tan Seok Shin . Maternal dietary patterns, diet quality and micronutrient status in gestational diabetes mellitus across different economies: A review. AIMS Medical Science, 2019, 6(1): 76-114. doi: 10.3934/medsci.2019.1.76
[6]	Salma M. AlDallal . Quick glance at Fanconi anemia and BRCA2/FANCD1. AIMS Medical Science, 2019, 6(4): 326-336. doi: 10.3934/medsci.2019.4.326
[7]	Houda Elfane, Sanaa El-Jamal, Mohamed Mziwira, Imane Barakat, Mohammed Elayachi, Rekia Belahsen . Nutritional status of Moroccan pregnant women and the birth weight of their newborn: a case-control study in the province of El Jadida. AIMS Medical Science, 2021, 8(3): 237-251. doi: 10.3934/medsci.2021020
[8]	Sylvia Kirchengast . Diabetes and Obesity—An Evolutionary Perspective. AIMS Medical Science, 2017, 4(1): 28-51. doi: 10.3934/medsci.2017.1.28
[9]	Hidetaka Hamasaki, Yasuteru Hamasaki . Nuts for Physical Health and Fitness: A Review. AIMS Medical Science, 2017, 4(4): 441-455. doi: 10.3934/medsci.2017.4.441
[10]	Hicham Rahmi, Ben Yamine Mallouki, Fatiha Chigr, Mohamed Najimi . The effects of smoking Haschich on blood parameters in young people from the Beni Mellal region Morocco. AIMS Medical Science, 2021, 8(4): 276-290. doi: 10.3934/medsci.2021023

Abstract

1. Introduction

Bitcoin markets are constantly influenced by new information (Karalevicius, 2018; Vidal-Tomás and Ibañez, 2018). They are actively traded 24 hours a day, seven days a week, and traders are heterogenous and need to adapt to ever-changing market conditions. While asset pricing theory states that all market participants act on the basis of the same information (Fama, 1970), in reality information is not equally distributed. Traders receive information at different times or interpret events and information differently. For this reason, information asymmetries and the dissemination of information need to be investigated in order to gain a better understanding of Bitcoin markets. The underlying transparency of the Bitcoin network offers the possibility to track any transfers on the public blockchain infrastructure virtually in real-time. As a result, no network participant is able to move large values without the market recognizing this.

However, the time of disclosure differs across market participants. In principle, there is always at least one party involved in a transaction, the initiator, who possesses private information in advance about a future transfer. Also, the verification of a transactions is not the first time that information about a transfer becomes publicly assessible. Once a transfer is initiated and sent to a node in the Bitcoin network, the node checks the transfer and forwards it to eight other nodes in the network, although the number of forwarding nodes is freely adjustable. The transaction is placed in the Mempool, a sort-of waiting area for pending transactions, and is picked up by miners who insert the transaction in their candidate blocks. Only once a miner successfully mines a transaction, the transfer is verified and becomes part of the blockchain. As on average, every ten minutes a new Bitcoin block is mined, nodes in the network possess private information about an upcoming transfer earlier than non-node market participants, although various risks are tied to unverified transactions, like double-spending attacks or network forks (Decker and Wattenhofer, 2016; Gervais et al., 2015).

While research has extensively analyzed the market economics and efficiency, as well as asset valuation and price formation of cryptocurrencies (Bariviera, 2017; Ciaian et al., 2016; Kristoufek, 2018; Li et al., 2018; Nadarajah and Chu, 2017; Sapuric et al., 2020), the relationship between cryptocurrency markets and blockchain transaction activity remains under-explored. This may be due to the fact that there are hardly any comparable phenomena in other financial markets, and thus existing scientific basis. So-called on-chain transaction activity represents a specific piece of public information that is part of Bitcoin’s microstructure. Koutmos (2018) shows for daily data that blockchain transaction activity of Bitcoin (the cumulative number of transactions and the number of unique blockchain addresses) can explain Bitcoin prices and Ante and Fiedler (2020) find abnormal returns associated with a market reaction to large Bitcoin transfers using minute price data. They show that transaction size and presumed motives of the transactions can explain price effects. In Aalborg et al. (2019), it is analyzed how transaction volume and the number of unique Bitcoin addresses relate to return, volatility and trading volume. The authors identify that changes in Bitcoin addresses positively relate to weekly and daily returns. While few studies have analyzed price effects and Bitcoin on-chain activity, the relationship between on-chain information (asymmetry) and trading volume remains unclear.

We are interested in the question if a specific type of unforeseen event on the Bitcoin blockchain, a large transfer, which we define as 500 or more Bitcoins being transacted, leads to a market reaction of abnormal trading activity and what factors may explain this behavior. For most transfers, the underlying motives remain mostly unclear, i.e. information asymmetry is high, while one or few involved actors possess private information. Therefore, uninformed market participants may not be willing to trade until the information asymmetry is resolved, which would result in a decrease in trading volume until the asymmetry is resolved (Chae, 2005). For some transfers, like intra-cryptocurrency exchange transfers, where Bitcoins are only moved for safekeeping, motives are for the most part clear, why such a large transfer could potentially have no influence on trading volume.

The aim of this study is to assess the statistical properties of actual minutely Bitcoin trading volume and abnormal trading volume for a sample of 2132 large Bitcoin transfers. We apply event study methodology to identify abnormal trading volumes and analyze if publicly known addresses of cryptocurrency exchanges, which can be monitored by market participants, can explain such effects. The results serve to better understand the link between Bitcoin’s on-chain activity and secondary market activity and to identify if trading volume of Bitcoin is linked to fundamental economic values and uncertainty. The findings contribute to the research on the market economics and efficiency of cryptocurrencies, which has been identified as a major scientific discourse (Ante, 2020). It also serves to better understand the behavior of participants in the Bitcoin market in response to unforeseen events and information asymmetry and shows how certain transactions, where senders or receivers can be identified, are interpreted by the market.

2. Conceptual background and hypotheses

2.1. Trading volume and its relation to unscheduled events

The opportunity for trade occurs when a potential buyer and seller change their price expectations. Reasons for this can be speculation or liquidity needs but are usually not assessible for external observers. If new information becomes known to the market that is relevant to the price expectations of market participants, it is priced in accordingly. The occurrence of abnormal trading volume reflects a change in the expectations of traders in a market due to an unforeseen or unusual event, which may lead to disagreement in interpretation of the information across investors (Beaver, 1968). However, it is also possible that the same interpretation of an event will lead to abnormal trading volumes, if ex-ante expectations differ across traders (Karpoff, 1986).

Trading volume represents a critical characteristic of financial markets, as it enables price discovery and financial risk-sharing. There are basically two forms of trading volume: informed and uninformed (liquidity trading). According to Kyle (1985), liquidity traders, like market makers, behave exogenously and inelastically to price, which should result in an increase in trading volume with increasing information asymmetry, as informed traders try to exploit private information (Chae, 2005). Another possibility, however, is that liquidity traders have timing discretion (Admati and Pfeiderer, 1988; Foster et al., 1984), which leads to a reduction of their trading volume with increasing information asymmetry. Liquidity traders react to exogenous trade demands by postponing trading until the information asymmetry is resolved (Chae, 2005). A higher probability to trade with informed counterparties leads to a decrease in the willingness of uninformed traders to participate in a market (Black, 1986; Milgrom and Stokey, 1982). So if uninformed traders suspect informed trading activities, which they oftentimes cannot (Easley et al., 2002), they should decrease their own trading activities. For (mostly) unscheduled events, like large Bitcoin transfers, uninformed investors cannot predict informed trading, i.e. they cannot change their trading activities accordingly. In the Bitcoin market, the executing entity (the sender) has private information at first. Subsequently, operators of Bitcoin nodes (e.g. miners) possess information about unconfirmed transactions until the confirmed transaction is finally broadcasted in a block (on average ten minutes later). Upon confirmation of the transaction, another group of market participants—those who observe the market but do not operate a node—receive information about the transaction, while another group receives information about the transaction with even more delay or not at all. To this extent, market participants can adjust their expectations of the Bitcoin market at different times and react accordingly. If large Bitcoin transfers do represent a relevant aspect of Bitcoin’s microstructure, trading volume should increase around transfers, as informed traders update their expectations and carry out purchases and/or sales related to it, while uninformed traders keep their volume at an usual level, resulting in a growth in total trading volume before and with large Bitcoin transactions.

HYPOTHESES 1: Trading volume increases before and with large Bitcoin transfers.

2.2. Bitcoin addresses and information asymmetry

While the Bitcoin blockchain is transparent in terms of verified transaction timing, size and (pseudo-anonymous) destination, one can only speculate about reasons for transactions. Basically, it can be said that a very large transaction between unknown addresses will lead to uncertainty, as one motive could be to sell the Bitcoins on a cryptocurrency exchange, which would lead to a negative price effect. Yet, different traders that observe such an event, may not agree on how to interpret it, i.e. how to adjust their price expectations and behavior. In addition, there are always one or two parties who are directly involved in the transfer and therefore have no or limited information asymmetry.

The Bitcoin addresses of cryptocurrency exchanges are often public knowledge, which allows the market to interpret transactions differently based on addresses they are sent to or initiated from. If one or two addresses involved in a transfer are known, information asymmetry is decreased and the interpretation of an unforeseen event with regard to market or price expectation may be less complex for traders. Exchanges usually use so-called hot wallets to handle daily business, such as deposits and withdrawals from customers. Only a relatively small proportion of all Bitcoins deposited on the exchange is stored on hot wallets—a proportion that is sufficient to guarantee daily payouts. The majority of an exchanges’ Bitcoins are stored on cold wallet addresses, which have significantly higher security standards and are only used rarely. For example, the cryptocurrency exchange Bitfinex states that about 99.5% of all user funds are in their offline, multi-signature wallet. Four out of seven hardware security modules in the possession of globally-distributed team members are required to initiate a transaction from the address (Bitfinex, 2020). In theory, market participants should know the addresses of major exchanges and may therefore interpret associated transactions differently than ones related to unknown addresses. In total, four overall transaction types based on Bitcoin addresses of cryptocurrency exchanges can be identified:

(1) Hot wallet deposits. As Bitcoins are sent to an exchange, the receiving entity is publicly known, i.e. information asymmetry is decreased. Traders are able to interpret possible transfer motivation, for example that Bitcoins are to be sold, shall be used as collateral for margin trading or are deposited to access services like insurance of funds.

(2) Hot wallet withdrawals. The initiating entity is known, as Bitcoins are withdrawn from the exchange to an unknown address.

(3) Cold wallet deposits. Excess Bitcoins are sent from the hot wallet of an exchange to the cold wallet, which can be interpreted as a sign that either deposits to the exchange increased or initiated withdrawals decreased. The associated level of information asymmetry is low, as (1) both sender and receiver are known and represent the same entity and (2) the likely transfer motive is known and should not affect the market directly.

(4) Cold wallet withdrawals. Excess Bitcoins are sent from the cold wallet of an exchange to the hot wallet, which can be interpreted as a sign that initiated withdrawals increased. Again, information asymmetry is low, as both involved addresses belong to the same entity and the transfer is company-internal.

In summary, transactions on the Bitcoin network can be identified where neither receiver nor initiator is known, where either sender or receiver is known or where both parties are known. The more related parties can be assigned to a transaction, the lower the level of information asymmetry. A lower level of information asymmetry should therefore lead to less disagreement regarding the interpretation of events, and thus lower abnormal trading volume once the information becomes known to the market (Beaver, 1968; Karpoff, 1986).

HYPOTHESES 2: The degree of information asymmetry tied to large transfers of Bitcoin is negatively correlated with trading volume after information becomes public knowledge.

3. Data and methods

3.1. Data and sample

Our data set consists of 2132 individual Bitcoin transactions where 500 or more Bitcoins were transferred. Blockchain data was collected from Bitcoin blockchain explorer blockchain.com for transactions that occurred between September 2018 and November 2019. We collected a transaction’s timestamp, its size, as well as sender and receiver addresses involved. This way, we are able to identify addresses that are repeatedly involved in transfers. Based on the sender and receiver addresses, we cluster transactions by publicly known Bitcoin addresses of cryptocurrency exchanges and the fact if assets are sent or received, i.e. deposits and withdrawals. For most cryptocurrency exchanges, information on Bitcoin wallets is publicly assessible, for example via bitinfocharts.com. For an overview of the clusters, see Table A.1. in the appendix.

To complement the data set, we collected price and volume data from the cryptocurrency exchange Gemini (gemini.com) for t = −200 to 200 minutes around the minute of each Bitcoin transfer. The trading volume of cryptocurrencies is oftentimes inflated or faked, which makes it difficult to analyze. The choice of Gemini as data basis allows to minimize this risk. The US-based exchange is (1) highly regulated, as it (a) has been awarded a BitLicense by the New York State Department of Financial Services, (b) is registered as a Money Services Business with FinCEN and (c) its funds are insured to a certain degree, (2) is a verified entity at the Blockchain Transparency Institute (bti.live), an industry-lead initiative for the provision of “clean and wash free data” and (3) did not show signs for wash trading in existing studies (Alameda Research, 2019; Fusaro and Hougan, 2019). The market data collected are close prices and trading volume per minute.

3.2. Dependent variables and event study methodology

Based on similar metrics used in the literature on trading volume in stock markets, two volume measures are used: (1) the number of Bitcoins traded per minute (e.g. Harris, 1986; James and Edmister, 1983) and (2) the dollar value of Bitcoins traded per minute (e.g. James and Edmister, 1983; Lakonishok and Vermaelen, 1986). Since both variables are skewed and minutes where no trading takes place appear in the data set, i.e. zero-inflated volume observations, we use the log (x + c) transformation of both variables, where x is the variable’s natural expression and c is a constant. The use of log transformed variables is recommended by the literature on abnormal trading volume (Ajinkya and Jain, 1989; Cready and Ramanan, 1991). In line with the literature on stock trading volume, we use a constant value of 1 for trading volume in USD (Ajinkya and Jain, 1989) and a value of 0.000255 for trading volume in the asset itself (Campbell and Wasley, 1996).

We build on volume event study methodology for our analysis. In event studies, the market reaction to specific unexpected or unusual information is assessed by adjusting observed market effects by expected market performance to identify abnormal effects. We choose a simple expectation model, where expected volume is calculated as the average volume over a 121-minutes estimation period (t = −141 to −21). The expected trading volume (EV_it) is calculated as the average volume over an estimation period as

$ {EV}_{it} = \overline{{V}_{it}}+{e}_{it} $

(1)

where i is the number of transactions and t is the number of minutes in the estimation period. The term V_it is the respective absolute trading volume, either in Bitcoin or in US-dollar, in minute t and transaction i, and e_it is the error term. The bar on top of the variable indicates the mean over the estimation window.

Abnormal trading volume ($ {\mathrm{A}\mathrm{V}}_{\mathrm{i}\mathrm{t}} $) can then be calculated by subtracting the observed trading volume from the expected trading volume:

$ {AV}_{it} = {V}_{it}-{EV}_{it} $

(2)

Across multiple events of the same type, i.e. transactions, AV_it can be agammaegated as average abnormal trading volume:

$ {AAV}_{it} = \frac{1}{N}\sum\limits _{i = 1}^{N}{AV}_{it} $

(3)

When assessing event windows beyond a single minute, the results are summarized in cumulative abnormal trading volume (CAV):

$ CAV({t}_{1}, {t}_{2}) = \sum\limits _{t = {t}_{1}}^{{t}_{2}}{AV}_{it} $

(4)

Cumulative average abnormal trading volume (CAAV_it) are calculated as the sum of the CAVs of various transactions:

$ {CAAV}_{it} = \frac{1}{N}\sum\limits _{i = 1}^{N}CAV({t}_{1}, {t}_{2}) $

(5)

To test the significance of abnormal trading volumes, we calculate parametric t-tests and the non-parametric Wilcoxon sign rank test (Wilcoxon, 1945), as such financial data is characterized by non-normality in its distribution (Brown and Warner, 1985).

3.3. Independent and control variables

We create five different dummy variables, each of which take a value of one if a transaction can be assigned to one of five clusters, and else zero. The variable hot wallet withdrawals indicates that a transaction originates from the hot wallet of a cryptocurrency exchange and is transferred to an address that does not belong to the exchange itself, i.e. the transaction is not sent to the cold wallet of the exchange. Analogously, the variable hot wallet deposits takes a value of one if Bitcoins are sent to the hot wallet of an exchange and the transfer is not initiated from the exchanges’ cold wallet. Cold wallet withdrawals indicates that a transaction is initiated from a cold wallet address of an exchange to any other address (usually the respective hot wallet of the exchange). The variable cold wallet deposits is assigned a value of one if Bitcoins are sent to the cold wallet of an exchange (again, usually from the respective hot wallet of the exchange).

As the transfers range from 500 Bitcoins upwards, it is conceivable that as the size of the transactions increases, effects will also increase. The variable transaction size (BTC) is calculated as the logarithm of the number of Bitcoins transferred.

Bitcoin is a volatile asset, whose value in dollars fluctuated strongly over the period under consideration from September 2018 to November 2019. A transfer of 500 or more Bitcoins may have very different equivalent value based on market timing. For this reason, we introduce the control variable Bitcoin price (＄1,000), which equals the US-dollar value of Bitcoin at the minute of the transaction, divided by 1000.

Trading volume is no stable mechanism but varies across daily hours and week days. Stock market studies show that average trading volume significantly differs for time-of-day and day-of-week (e.g. Jain and Joh, 1988). The assessment of day-of-week effects for Bitcoin has shown that returns are higher on Mondays (Caporale and Plastun, 2019), lower on Sundays (Dorfleitner and Lung, 2018) and trading activity is lower on weekends, as e.g. institutional traders do not trade on weekends (Baur et al., 2019; Kaiser, 2019; Wang et al., 2019). To assess and control for day-of-week effects, we create seven dummy variables, each taking the value of one if the transaction happened on a specific weekday, and else zero. Analogously we create a dummy variable for each hour of the day to control for time-of-day effects.

4. Results

4.1. Descriptive statistics

Table 1 reports summary statistics per month for the sample of large transfers and the Bitcoin market. We identified three-digit numbers of large transfers for the first nine months and lower numbers thereafter. The number of large transfers seems to be negatively correlated to the four metrics total transactions (on the blockchain), price, volume and market capitalization—a plausible fact, as traders seem to orientate themselves at the dollar value of the asset. The price of Bitcoin dropped from ＄6610 to $3701 over the course of five months before rising to ＄10,669 over the next six months, “ending” at ＄8373. This high volatility over the period where data on large transfers has been collected suggests that price should be included as a control variable in empirical models.

Table 1. Sample distribution and Bitcoin statistics by month.

Month	Large Bitcoin transfers(sample)	Bitcoin transactions (million)	Bitcoinprice (＄)	Bitcoin trading volume(＄ billion)	Bitcoin market capitalization(＄ billion)
Sep 18	135	68.71	6,610.68	4.325	114.171
Oct 18	195	78.63	6,485.12	3.821	112.360
Nov 18	110	80.15	5,404.25	5.279	93.901
Dec 18	403	81.91	3,717.49	5.447	64.787
Jan 19	168	93.25	3,701.56	5.398	64.720
Feb 19	199	92.87	3,711.91	7.111	65.106
Mar 19	128	91.04	3,976.07	9.666	69.882
Apr 19	269	110.03	5,178.47	14.845	91.394
May 19	176	115.05	7,309.70	23.360	129.436
Jun 19	84	104.86	9,415.90	22.529	167.272
Jul 19	63	103.05	10,669.34	21.820	190.132
Aug 19	50	101.7	10,643.25	17.225	190.297
Sep 19	51	97.99	9,814.07	16.018	176.041
Oct 19	73	102.09	8,411.93	19.200	151.386
Nov 19	28	92.02	8,373.57	22.564	151.135
Note: Price, trading volume and market capitalization data has been extracted from crypto data aggregator coinmarketcap.com and are calculated as means over the daily metrics of each month. Bitcoin transaction data has been collected from blockchain.info.

| Show Table

DownLoad: CSV

Table 2 reports summary statistics for the two volume measures, both the untransformed and the log-transformed expressions for all transactions, i.e. the full data set, the five address clusters and deciles based on the value in USD of a Bitcoin transfer. Mean and standard deviation statistics on trading volume are calculated as the average over the 31-minute-long event window starting 15 minutes before the event. The trading volume per minute for the 31-minute event window and the 401-minute data sample are visualized in Figures A.1 through A.3 for trading volume in USD, both for the full set of transactions and across address clusters and deciles.

Table 2. Summary statistics of trading volume per minute with and without log transformation around 2132 large Bitcoin transactions between September 2018 and November 2019.

		Mean valuetransacted		Untransformed trading volume measures				Log-transformed trading volume measures
				Number of Bitcoins traded		USD value traded		Number of Bitcoins traded		USD value traded
	N	USD (million)	Bitcoin	Mean	(SD)	Mean	(SD)	Mean	(SD)	Mean	(SD)
All transactions	2,132	12.45	2,029	3.081	(11.36)	0.174	(0.744)	−2.605	(3.844)	5.900	(3.939)
Non-exchange transfers	434	21.09	2,745	3.468	(12.74)	0.238	(1.012)	−2.575	(3.759)	6.054	(3.898)
Hot wallet deposits	612	8.25	1,565	3.406	(13.91)	0.173	(0.826)	−2.416	(3.877)	6.008	(3.939)
Hot wallet withdrawals	906	8.53	1,531	2.894	(9.438)	0.153	(0.568)	−2.623	(3.857)	5.857	(3.944)
Cold wallet deposits	51	12.04	1,920	2.267	(5.266)	0.145	(0.345)	−2.661	(3.701)	5.970	(3.850)
Cold wallet withdrawals	129	31.02	5,372	1.869	(5.707)	0.117	(0.367)	−3.451	(3.867)	5.136	(4.045)
Lowest decile	214	1.96	520	4.747	(13.12)	0.182	(0.476)	−1.360	(3.691)	6.866	(3.686)
2	212	3.29	730	2.274	(8.084)	0.113	(0.382)	−3.325	(3.965)	5.114	(4.059)
3	212	4.19	919	2.698	(8.584)	0.127	(0.420)	−2.701	(3.859)	5.753	(3.920)
4	212	5.14	1,182	2.783	(14.33)	0.161	(1.023)	−2.559	(3.743)	5.826	(3.811)
5	216	5.90	1,318	2.233	(5.798)	0.097	(0.262)	−2.511	(3.807)	5.856	(3.824)
6	213	6.46	1,298	2.696	(13.12)	0.120	(0.479)	−2.508	(3.834)	6.014	(3.906)
7	215	7.64	1,497	4.713	(15.94)	0.290	(1.180)	−2.470	(3.884)	6.014	(3.986)
8	211	10.27	1,555	3.413	(12.14)	0.217	(0.845)	−2.469	(3.724)	6.298	(3.883)
9	214	14.07	2,126	1.639	(5.930)	0.154	(0.664)	−3.405	(3.799)	5.265	(4.039)
Largest decile	213	65.62	9,091	3.592	(11.39)	0.280	(1.035)	−2.761	(3.846)	5.977	(4.076)
Note: Untransformed USD value traded is divided by 100,000 for readability. Deciles are calculated based on USD value of the Bitcoin transfers. Mean statistics on trading volume are calculated as minute averages over the 31-minute full event window centered around each Bitcoin transaction (t = −15 to 15).

| Show Table

DownLoad: CSV

For the full set of transactions, the untransformed variables have high skewness (14.12 for Bitcoin; 8.76 for USD) and kurtosis (255.19 for Bitcoin; 108.14 for USD), which clearly shows that there is no normality in distribution. As described above, we use log-transformed variables for correction, which clearly improves skewness (−0.35 for Bitcoin; −0.39 for USD) and kurtosis (0.82 for Bitcoin; 0.86 for USD). The variables’ distributions are only close to normality, why, in addition to the t-test, the non-parametric Wilcoxon sign rank test is also used to test for significance.

The average trading volume per minute around large Bitcoin transfers is ＄17,400 (or 3.081 Bitcoins), which would—in case there are not any abnormal effects identified—result in a daily traded value of little over ＄25 million (for the cryptocurrency exchange under consideration). The mean traded volume is highest for the sample of non-exchange transfers (＄23,800), i.e. the proxy variable with the highest level of associated information asymmetry, and lowest for the two cold wallet variables (＄11,700–14,500), i.e. the proxies with lowest associated information asymmetry. Standard deviations are also highest for non-exchange transfers (＄101,200) and lowest for cold wallet transactions (＄34,500–36,700).

For the size-based deciles, we identify, in comparison to the other lower deciles, a high level of associated trading volume in the lowest decile (＄18,200). For three deciles (7th, 8th and 10th), trading volumes higher than ＄20,000 per minute are identified. This shows that trading volume seems to increase with the size of transactions, but size does not seem to be the only decisive factor.

4.2. Empirical results

4.2.1. Abnormal trading volume around large Bitcoin transactions

Abnormal trading volumes for both trading volume measures described in the previous sections are reported in Table 3 for 21 individual minutes around the Bitcoin transactions (t = −10 to 10) and for four different event windows, two for minutes leading up to the transfer and two for periods starting with the minute of the transfer. We identify significant positive abnormal trading volume for minutes leading up to the transaction, for the minute where the transaction is verified and for minutes thereafter. Both event windows (t = −15 to −1 and −5 to −1) before the Bitcoin transfer is verified show highly significant positive abnormal trading volume. The results obtained therefore allow hypothesis 1 to be confirmed.

Table 3. Abnormal trading volumes around large Bitcoin transactions.

Minute	Volume in USD				Volume in Bitcoin
Minute	AAV	t-test	z-test	positive	AAV	t-test	z-test	positive
−10	0.1992	2.61^***	3.63^***	57%	0.1895	2.54^**	3.40^***	56%
−9	0.0600	0.79	1.76^*	56%	0.0529	0.71	1.56	55%
−8	−0.0451	−0.58	0.26	55%	−0.0474	−0.63	0.10	54%
−7	0.0466	0.61	1.43	54%	0.0352	0.47	1.19	54%
−6	0.0373	0.48	1.30	54%	0.0288	0.38	1.07	53%
−5	0.2459	3.28^***	4.27^***	57%	0.2307	3.16^***	4.02^***	57%
−4	0.1497	1.99^**	2.97^***	56%	0.1388	1.89^*	2.73^***	55%
−3	0.1176	1.54	2.36^**	55%	0.1038	1.39	2.09^**	54%
−2	0.1298	1.70^*	2.84^***	57%	0.1176	1.58	2.60^***	56%
−1	0.0430	0.56	1.75^*	56%	0.0318	0.42	1.54	56%
0	0.1905	2.50^**	3.46^***	56%	0.1722	2.31^**	3.19^***	56%
1	0.0453	0.59	1.37	54%	0.0320	0.43	1.11	54%
2	0.1447	1.89^*	2.78^***	56%	0.1302	1.75^*	2.51^**	55%
3	0.1022	1.35	2.30^**	57%	0.0860	1.16	2.02^**	56%
4	0.1505	1.97^**	2.65^***	55%	0.1352	1.81^*	2.39^**	55%
5	0.1547	2.02^**	3.14^***	56%	0.1435	1.92^*	2.93^***	56%
6	−0.0142	−0.17	0.60	54%	−0.0162	−0.21	0.41	54%
7	0.1214	1.55	2.63^***	57%	0.1170	1.54	2.46^**	56%
8	0.1442	1.85^*	3.01^***	57%	0.1336	1.77^*	2.80^***	56%
9	0.1897	2.37^**	3.21^***	58%	0.1818	2.33^**	3.07^***	57%
10	0.1296	1.63	2.40^**	55%	0.1244	1.61	2.23^**	55%
Window	CAV	t-test	z-test	positive	CAV	t-test	z-test	positive
[−15, −1]	1.6316	2.99^***	2.29^**	51%	1.4878	2.79^***	2.08^**	51%
[−5, −1]	0.6860	2.91^***	2.60^***	52%	0.6227	2.71^***	2.37^**	52%
[0, 5]	0.7878	2.92^***	1.96^**	52%	0.6992	2.66^***	1.70^*	52%
[0, 15]	1.5745	2.58^**	0.50	49%	1.4159	2.39^**	0.28	49%
Note: , , ** indicates significance at the 10%, 5% and 1% level, respectively. The column z-test refers to the non-parametric Wilcoxon sign rank test. The column ‘positive’ shows the share of observations with positive abnormal trading volume for the respective period.

| Show Table

DownLoad: CSV

The results show that the market is interpreting information on large Bitcoin transfers before they are verified. Therefore, informed traders in the form of e.g. involved entities in transactions, node operators or mining ventures try to exploit private information. Uninformed traders are not able to identify that informed counterparties participate in the market (Easley et al., 2002) and therefore do not decrease their own trading.

When comparing the two different volume measures, we identify a high degree of similarity between them, an understandable fact, as both measures show a significant correlation coefficient of 0.94. Due to the high correlation and the identified similarities, we will only examine volumes in USD as a metric in the following analysis, as significance levels are slightly higher and the metric has higher relevance.

4.2.2. Abnormal trading volume and effects of information asymmetry

We analyze the causes of abnormal trading volume around unscheduled large Bitcoin transfers by regressing abnormal returns on information asymmetry proxy variables and control variables. As information asymmetry cannot be measured directly, we use the address cluster dummy variables as proxies. Intuitively, a greater level of information asymmetry should lead to lower levels of trading. Thus, non-exchange transfers, where both sender and receiver are unknown should have the highest, while cold wallet deposits and withdrawals should have the lowest effects. We control for transaction size (BTC), Bitcoin price (＄1,000), day-of-week effects and time-of-day effects. The results are shown in Table 4. Descriptive statistics and correlations of the main variables can be assessed in Tables A.2. and A.3. in the appendix. Statistics with further information on day-of-week effects (Table A.4) and time-of-day effects (Figures A.4 and A.5) can also be found in the annex. The daily effects confirm previous findings on the phenomenon in cryptocurrency markets that were described in section 3.3 (Baur et al., 2019; Caporale and Plastun, 2019; Dorfleitner and Lung, 2018; Kaiser, 2019; Wang et al., 2019). In the previous analysis, highly significant positive abnormal trading volumes were identified for the periods t = −15 to −1, t = 0 and t = 0 to 5. Therefore, the abnormal trading volumes of these three periods are chosen as dependent variables for regression analysis. This way, we can assess effects of asymmetric information before, at and after events.

Table 4. Regression models predicting average abnormal trading volume in USD across different periods.

Independent variables (information asymmetry proxies) and control variables
Non-exchangetransfers	Hot walletwithdrawals	Hot walletdeposits	Cold walletwithdrawals	Cold walletdeposits	Bitcoin price(in ＄1,000)	Transactionsize (BTC)	R²
Panel A: Abnormal trading volume in t = −15 to −1
3.981 (1.442)^***					−0.397 (0.225)^*	0.689 (0.814)	0.045
	1.122 (1.122)				−0.255 (0.219)	0.546 (0.809)	0.043
		−4.696 (1.243)^***			−0.420 (0.221)^*	0.569 (0.807)	0.049
			0.739 (2.287)		−0.273 (0.219)	0.471 (0.811)	0.042
				0.396 (4.076)	−0.268 (0.219)	0.511 (0.810)	0.042
2.428 (2.497)	−0.268 (2.370)	−4.297 (2.475)^*		−0.158 (4.633)	−0.488 (0.227)^**	0.666 (0.812)	0.050
Panel B: Abnormal trading volume in t = 0
−0.367 (0.200)^*					0.059 (0.035)^*	−0.071 (0.108)	0.021
	0.092 (0.158)				0.048 (0.034)	−0.052 (0.108)	0.020
		−0.283 (0.175)^*			0.038 (0.034)	−0.051 (0.108)	0.021
			1.199 (0.346)^***		0.036 (0.034)	−0.121 (0.109)	0.025
				0.920 (0.447)^**	0.046 (0.034)	−0.058 (0.108)	0.021
−1.391 (0.379)^***	−1.079 (0.357)^***	−1.327 (0.370)^***		−0.232 (0.551)	0.039 (0.035)	−0.129 (0.109)	0.029
Panel C: Abnormal trading volume in t = 0 to 5
−1.143 (0.694)^*					0.266 (0.116)^**	0.452 (0.367)	0.045
	0.525 (0.544)				0.233 (0.114)^**	0.517 (0.368)	0.044
		−0.794 (0.603)			0.202 (0.115)^*	0.513 (0.368)	0.045
			3.246 (1.171)^***		0.202 (0.114)^*	0.324 (0.373)	0.048
				0.800 (2.172)	0.227 (0.114)^**	0.500 (0.369)	0.044
−3.891 (1.287)	^***	−2.746 (1.213)^**	−3.586 (1.253)^***	−2.281 (2.438)	0.323 (0.114)^***	0.309 (0.371)	0.049
Note: N = 2132; , , ** indicates significance at the 10%, 5% and 1% level, respectively. Standard errors are robust to heteroskedasticity. Cold wallet withdrawals are the reference group in each panels’ final model. Constant term, day-of-week and time-of-day control variables are suppressed.

| Show Table

DownLoad: CSV

For the period leading up to transfers, we identify that non-exchange transfers positively affect abnormal trading volumes, although the effect does not hold in the model where all information asymmetry proxies are tested. Hot wallet deposits have significant negative coefficients in both tested models, while all other proxy variables lack significance. The results indicate that informed traders interpret initiated large transfers differently. At the minute the transaction is verified and the event become public knowledge, non-exchange transfers show the highest significant negative effect on abnormal volume, while the coefficients for hot wallet deposits are also negative and significant.

The effect of hot wallet withdrawals is significantly negative for the model where all information asymmetry proxies are included. The observed effects turn for transactions where exchange cold wallets are included: Both coefficients for cold wallet deposits and withdrawals show significant and positive effects on abnormal trading volume in the individual models.

The same type of effect characteristics can be observed for the six-minute period from the public announcement of the confirmed transaction. Non-exchange transfers have highest negative effects, followed by hot wallet transfers, and cold wallet withdrawals positively affect volume, while cold wallet deposits lack significance.

The results indicate that, in line with hypotheses 2, once information about transfers becomes public knowledge, the market assesses the level of information asymmetry tied to the transfers and reacts accordingly. A higher level of information asymmetry results in lower levels of abnormal trading, as uninformed traders’ probability to trade with informed traders increases and thus their willingness to participate in the market decreases (Black, 1986; Milgrom and Stokey, 1982).

The Bitcoin price is a significantly positive predictor of abnormal trading volume for the ex-post trading volume (panel c), and a negative—and in some models significant—predictor for ex-ante trading volume. This suggests that market sentiment, i.e. periods with higher Bitcoin prices, result in larger market reactions. This can either be an increase in trading of existing traders or an increase in the number of market participants.

6. Concluding remarks

Using data on large Bitcoin transfers, we find positive abnormal trading volume with large transfers on the Bitcoin blockchain. Trading volume increases even before transactions are confirmed by the network, which can be explained by the fact that informed traders, i.e. traders that operate a Bitcoin node, change their trading behavior based on the information as soon as they learn about an upcoming transaction. Uninformed traders, i.e. market participants who do not operate nodes, cannot identify that informed traders have changed their behavior on the basis of private information and therefore continue to trade at the same level.

Another result is that abnormal trading volume negatively correlates with the degree of information asymmetry associated with transactions, as has been shown for other financial markets (Chae, 2005). We identify this by using the involvement of publicly known Bitcoin addresses of cryptocurrency exchanges in the transactions as a proxy for information asymmetry. Transactions in which neither the initiator nor the receiver are known Bitcoin addresses show the largest positive effect on ex-ante abnormal trading volumes and the largest negative effect ex-post. Transfers of Bitcoin as relocation within a cryptocurrency exchange, i.e. where both initiator and receiver are public knowledge and the market can guess probable motives of a transfer, even positively effect ex-post trading volume.

The main implication of this study is that, as already identified for returns (Ante and Fiedler, 2020; Koutmos, 2018), on-chain activity is a relevant aspect of the microstructure of Bitcoin. The results indicate that specific traders make use of this type of private information, while at least some liquidity traders seem not to do, as they would use timing discretion and reduce their trading volume until the information asymmetry is resolved (Admati and Pfeiderer, 1988; Chae, 2005; Foster et al., 1984). As anyone is able to deploy a node in the Bitcoin network, liquidity traders should operate nodes themselves in order to close the information gap and to be able to assess the probability of trading with informed counterparties (Black, 1986)—yet, the initiator of a transaction will always remain as informed counterparty.

This also results in an open question for future studies. One could classify the moment when a transaction is identified by a node for the first time and the transaction is sent to the Mempool as the announcement date of a transaction. If abnormal trading volumes continue to occur prior to the announcement of a transaction, the only explanation would be that parties directly involved in the transaction are using their private information. Further investigations could of course focus on a more detailed identification of known addresses in the network or investigate other cryptocurrencies. Studies investigating these issues will further serve to understand the relationship between blockchain network transactions and secondary markets of Bitcoin and other cryptocurrencies.

Acknowledgments

We would like to thank Elias Strehle for his help in collecting and processing the data.

Conflicts of interest

The author declares no conflict of interest in this paper.

References

[1]	Happé F (2015) Autism as a neurodevelopmental disorder of mind-reading. Abstr Book 3: 197–209.
[2]	Dover CJ, Le Couteur AL (2007) How to diagnose Autism. Arch Dis Child 92: 540–545. doi: 10.1136/adc.2005.086280
[3]	Jukić V, Arbanas G (2013) Diagnostic and Statistical Manual of Mental Disorders; Fifth Edition (DSM-5).Int J 57: 1546–1548.
[4]	Fombonne E (2009) Epidemiology of pervasive developmental disorders. Pediatr Res 65: 591–598. doi: 10.1203/PDR.0b013e31819e7203
[5]	Fombonne E, Quirke S, Hagen A (2009) Prevalence and interpretation of recent trends in rates of pervasive developmental disorders. McGill J Med 12: 73.
[6]	Weintraub K (2011) Autism counts. Nature 479: 22–24. doi: 10.1038/479022a
[7]	Nazeer A, Ghaziuddin M (2012) Autism spectrum disorders: Clinical features and diagnosis. Pediatr Clin North Am 59: 19–25. doi: 10.1016/j.pcl.2011.10.007
[8]	Zablotsky B, Black LI, Maenner MJ, et al. (2015) Estimated prevalence of autism and other developmental disabilities following questionnaire changes in the 2014 national health interview survey. Natl Health Stat Rep 2015: 1–20.
[9]	Hansen SN, Schendel DE, Parner ET (2015) Explaining the increase in the prevalence of autism spectrum disorders the proportion attributable to changes in reporting practices. JAMA Pediatr 169: 56–62. doi: 10.1001/jamapediatrics.2014.1893
[10]	Mostafa GA, ALayadhi LY (2012) Reduced serum concentrations of 25-hydroxy vitamin D in children with autism: Relation to autoimmunity. J Neuroinflammation 9: 1–7.
[11]	Duan XY, Jia FY, Jiang HY (2013) Relationship between vitamin D and autism spectrum disorder. Chin J Contemp Pediatr 15: 698–702.
[12]	Newschaffer CJ, Croen LA, Daniels J, et al. (2007) The epidemiology of autism spectrum disorders. Annu Rev Public Health 28: 235–258. doi: 10.1146/annurev.publhealth.28.021406.144007
[13]	Christison GW, Ivany K (2006) Elimination diets in autism spectrum disorders: Any wheat amidst the chaff? J Dev Behav Pediatr 27: S162–S171. doi: 10.1097/00004703-200604002-00015
[14]	Meyer U, Feldon J, Dammann O (2011) Schizophrenia and autism: Both shared and disorder-specific pathogenesis via perinatal inflammation? Pediatr Res 69: 26R–33R.
[15]	Wagner CL, Taylor SN, Dawodu A, et al. (2012) Vitamin D and its role during pregnancy in attaining optimal health of mother and fetus. Nutrients 4: 208–230.
[16]	Abdulbari B, Oaa AHA, Saleh NM (2013) Association between vitamin D insufficiency and adverse pregnancy outcome: Global comparisons. Int J Womens Health 5: 523.
[17]	Georgieff MK (2007) Nutrition and the developing brain: Nutrient priorities and measurement. Am J Clin Nutr85: 614S–620S.
[18]	Al-Farsi YM, Waly MI, Deth RC, et al. (2013) Impact ofnutrition on serum levels of docosahexaenoic acid among Omani children with autism. Nutrition 29: 1142–1146.
[19]	Bell JG, Mackinlay EE, Dick JR, et al. (2004) Essential fatty acids and phospholipase A2 in autistic spectrum disorders. Prostaglandins Leukotrienes Essent Fatty Acids 71: 201–204.
[20]	Amminger GP, Berger GE, Schafer MR, et al. (2007) Omega-3 fatty acids supplementation in children with autism: A double-blind randomized, placebo-controlled pilot study. Biol Psychol 61: 551–553.
[21]	Meguid NA, Atta HM, Gouda AS, et al. (2008) Role of polyunsaturated fatty acids in the management of Egyptian children with autism. Clin Biochem 41: 1044–1048.
[22]	Meiri G, Bichovsky Y, Belmaker RH (2009) Omega 3 fatty acid treatment in autism. J Child Adolesc Psychopharmacol 19: 449–451.
[23]	El-Ansary AK, Ben BAG, Al-Ayahdi LY (2011) Impaired plasma phospholipids and relative amounts of essential polyunsaturated fatty acids in autistic patients from Saudi Arabia. Lipids Health Dis 10: 63. doi: 10.1186/1476-511X-10-63
[24]	Yui K, Koshiba M, Nakamura S, et al. (2012) Effects of large doses of arachi-donic acid added to docosahexaenoic acid on social impairment in individuals with autism spectrum disorders: A double-blind, placebo-controlled, randomized trial. J Clin Psychopharmacol 32: 200–206.
[25]	Gómezpinilla F (2008) Brainfoods: The effect of nutrients on brain function. Nat Rev Neurosci 9: 568–578.
[26]	Willis LM, Shukitt-Hale BJ (2009) Recent advances in berry supplementation and age-related cognitive decline. Curr Opin Clin Nutr Metab Care 12: 91–94. doi: 10.1097/MCO.0b013e32831b9c6e
[27]	Gu Y, Nieves JW, Stern Y, et al. (2010) Food combination and Alzheimer disease risk: A protective diet. Arch Neurol 67: 699–706.
[28]	Nyaradi A, Li J, Hickling S, et al. (2013) The role of nutrition in children's neurocognitive development, from pregnancy through childhood. Front Hum Neurosci 7: 97.
[29]	Prado EL, Dewey KG (2014) Nutrition and brain development in early life. Nutr Rev 72: 267–284.
[30]	Keunen K, Elburg RMV, Bel FV, et al. (2014) Impact of nutrition on brain development and its neuroprotective implications following preterm birth. Pediatr Res 77: 148–155.
[31]	Georgieff MK, Rao R, (2001) The role of nutrition in cognitive development. In: Nelson CA, Luciana M, Eds.,Handbook in developmental cognitive neuroscience.Cambridge, MA: MIT Press, 491–504.
[32]	Hultman CM, Sparén P, Cnattingius S (2002) Perinatal risk factors for infantile autism. Epidemiology 13: 417–423.
[33]	Dionne G, Boivin M, Seguin JR. et al. (2008) Gestational diabetes hinders language development in offspring. Pediatrics 122: 1073–1079.
[34]	Leonard H, Klerk N, Bourke J, et al. (2006) Maternal health in pregnancy and intellectual disability in the offspring: A population-based study. Ann Epidemiol 16: 448–454.
[35]	Dodds L, Fell DB, Shea S, et al. (2011) The role of prenatal, obsteric and neonatal factors in the development of autism. J Autism Dev Disord 41: 891–902.
[36]	Burdge GC, Lillycrop KA (2014) Fatty acids and epigenetics. Curr Opin Clin Nutr Metab Care 17: 156–161. doi: 10.1097/MCO.0000000000000023
[37]	Delong GR (1993) Effects of nutrition on brain development in humans. Am J Clin Nutr 57: S286–S290.
[38]	Morgane PJ, Mokler DJ, Galler JR (2002) Effects of prenatal protein malnutrition on the hippocampal formation. Neurosci Biobehav Rev 26: 471–483. doi: 10.1016/S0149-7634(02)00012-X
[39]	Rosales FJ, Reznick JS, Zeisel SH (2009) Understanding the role of nutrition in the brain and behavioral development of toddlers and preschool children: Identifying and addressing methodological barriers. Nutr Neurosci 12: 190–202. doi: 10.1179/147683009X423454
[40]	Mccann JC, Ames BN (2005) Is docosahexaenoic acid, an n-3 long-chain polyunsaturated fatty acid, required for development of normal brain function? An overview of evidence from cognitive and behavioral tests in humans and animals. Am J Clin Nutr 82: 281–295.
[41]	Innis SM (2007) Dietary (n-3) fatty acids and brain development. J Nutr 137: 855–859. doi: 10.1093/jn/137.4.855
[42]	Wu A, Ying Z, Gomezpinilla F (2007) Omega-3 fatty acids supplementation restores mechanisms that maintain brain homeostasis in traumatic brain injury. J Neurotrauma 24: 1587–1595.
[43]	De Souza AS, Fernandes FS (2011) Effects of maternal malnutrition and postnatal nutritional rehabilitation on brain fatty acids, learning, and memory. Nutr Rev 69: 132–144.
[44]	Lozoff B, Georgieff MK (2006) Iron deficiency and brain development. Semin Pediatr Neurol 13: 158–165.
[45]	Zimmermann MB (2011) The role of iodine in human growth and development. Semin Cell Dev Biol 22: 645–652.
[46]	Greenwood CE, Winocur G (2005) High-fat diets, insulin resistance and declining cognitive function. Neurobiol Aging 26: 42–45.
[47]	Molteni R, Barnard RJ, Ying Z, et al. (2002) A high-fat, refined sugar diet reduces hippocampal brain-derived neurotrophic factor, neuronal plasticity, and learning. Neuroscience 112: 803–814.
[48]	Rice D, Barone S (2000) Critical periods of vulnerability for the developing nervous system: Evidence from humans and animal models. Environ Health Perspect 108: 511–533.
[49]	Couperus JW,Nelson CA,(2006) Early brain development and plasticity. In: McCartney K, Phillips D, Eds.,The Blackwell Handbook of Early Childhood Development.Malden, MA: Blackwell Publsihing,85–105.
[50]	Benton D (2010) The influence of dietary status on the cognitive performance of children. Mol Nutr Food Res 54: 457–470.
[51]	Levitsky DA, Strupp BJ (1995) Malnutrition and the brain: Changing concepts, changing concerns. J Nutr 125: 2212S–2220S. doi: 10.1093/jn/125.suppl_8.2212S
[52]	Penido AB, Rezende GH, Abreu RV, et al. (2012) Malnutrition during central nervous system growth and development impairs permanently the subcortical auditory pathway. Nutr Neurosci 15: 31–36. doi: 10.1179/1476830511Y.0000000022
[53]	Roseboom TJ, Painter RC, Van Abeelen AF, et al. (2011) Hungry in the womb: What are the consequences? Lessons from the Dutch famine. Maturitas 70: 141–145.
[54]	Kerac M, Postels DG, Mallewa M, et al. (2014) The interaction of malnutrition and neurologic disability in Africa. Semin Pediatr Neurol 21: 42–49. doi: 10.1016/j.spen.2014.01.003
[55]	Jacka FN, Pasco JA, Mykletun A, et al. (2010) Association of western and traditional diets with depression and anxiety in women. Am J Psychiatry 167: 305–311. doi: 10.1176/appi.ajp.2009.09060881
[56]	Jacka FN, Pasco JA, Mykletun A, et al. (2011) Diet quality in bipolar disorder in a population-based sample of women. J Affective Disord 129: 332–337. doi: 10.1016/j.jad.2010.09.004
[57]	Forsyth AK, Williams PG, Deane FP (2011) Nutrition status of primary care patients with depression and anxiety. Aust J Primary Health 18: 172–176.
[58]	Scarmeas N, Stern Y, Tang MX, et al. (2006) Mediterranean diet and risk for Alzheimer's disease. Ann Neurol 59: 912–921. doi: 10.1002/ana.20854
[59]	Francesco S, Francesca C, Rosanna A, et al. (2008) Adherence to Mediterranean diet and health status: A meta-analysis. BMJ 337: a1344. doi: 10.1136/bmj.a1344
[60]	Scarmeas N, Stern Y, Mayeux R, et al. (2009) Mediterranean diet and mild cognitive impairment. Arch Neurol 66: 216–225.
[61]	Eilander A, Gera T, Sachdev HS, et al. (2010) Multiple micronutrient supplementation for improving cognitive performance in children: Systematic review of randomised controlled trials. Am J Clin Nutr 91: 115–130. doi: 10.3945/ajcn.2009.28376
[62]	Schoenthaler SJ, Amos S, Doraz W, et al. (1997) The effect of randomized vitamin-mineral supplementation on violent and non-violent antisocial behavior among incarcerated juveniles. J Nutr Environ Med 7: 343–352. doi: 10.1080/13590849762475
[63]	Schoenthaler SJ, Bier ID (1999) Vitamin-mineral intake and intelligence: A macrolevel analysis of randomised controlled trials. J Altern Complementary Med 5: 125–134. doi: 10.1089/acm.1999.5.125
[64]	Zaalberg A, Nijman H, Bulten E, et al. (2010) Effects of nutritional supplements on aggression, rule-breaking, and psychopathology among young adult prisoners. Aggressive Behav 36: 117–126. doi: 10.1002/ab.20335
[65]	Buffington SA, Di PG, Auchtung TA, et al. (2016) Microbial reconstitution reverses maternal diet-induced social and synaptic deficits in offspring. Cell 165: 1762–1775. doi: 10.1016/j.cell.2016.06.001
[66]	Bazinet RP, Layé S (2014) Polyunsaturated fatty acids and their metabolites in brain function and disease. Nat Rev Neurosci 15: 771–785. doi: 10.1038/nrn3820
[67]	Innis S, De-La-Presa-Owens S (2001) Dietary fatty acid composition in pregnancy alters neurite membrane fatty acids and dopamine in newborn rat brain. J Nutr 131: 118–122. doi: 10.1093/jn/131.1.118
[68]	Pardo CA, Eberhart CG (2007) The neurobiology of autism. Brain Pathol 17: 434–447. doi: 10.1111/j.1750-3639.2007.00102.x
[69]	Chalon S (2006) Omega-3 fatty acids and monoamine neurotransmission. Prostaglandins Leukotrienes Essent Fatty Acids 75: 259–269. doi: 10.1016/j.plefa.2006.07.005
[70]	Zimmer L, Delpal S, Guilloteau D, et al. (2000) Chronic n-3 polyunsaturated fatty acid deficiency alters dopamine vesicle density in the rat frontal cortex. Neurosci Lett 284: 25–28. doi: 10.1016/S0304-3940(00)00950-2
[71]	Aïd S, Vancassel S, Poumès-Ballihaut C, et al. (2003) Effect of a diet-induced n-3 PUFA depletion on cholinergic parameters in the rat hippocampus. J Lipid Res 44: 1545–1551. doi: 10.1194/jlr.M300079-JLR200
[72]	Takeuchi T, Iwanaga M, Harada E (2003) Possible regulatory mechanism of DHA-induced anti-stress reaction in rats. Brain Res 964: 136–143. doi: 10.1016/S0006-8993(02)04113-6
[73]	Fedorova I, Alvheim AR, Hussein N, et al. (2009) Deficit in prepulse inhibition in mice caused by dietary n-3 fatty acid deficiency. Behav Neurosci 123: 1218–1225. doi: 10.1037/a0017446
[74]	Larrieu T, Hilal LM, Fourrier C, et al. (2014) Nutritional omega-3 modulates neuronal morphology in the prefrontal cortex along with depression-related behaviour through corticosterone secretion. Transl Psychiatry 4: e437. doi: 10.1038/tp.2014.77
[75]	Larrieu T, Madore C, Joffre C, et al. (2012) Nutritional n-3 polyunsaturated fatty acids deficiency alters cannabinoid receptor signalling pathway in the brain and associated anxiety-like behavior in mice. J Physiol Biochem 68: 671–681. doi: 10.1007/s13105-012-0179-6
[76]	Jones ML, Mark PJ, Waddell BJ (2013) Maternal omega-3 fatty acid intake increases placental labyrinthine antioxidant capacity but does not protect against fetal growth restriction induced by placental ischaemia-reperfusion injury. Reproduction 146: 539–547. doi: 10.1530/REP-13-0282
[77]	Li Q, Leung YO, Zhou I, et al. (2015) Dietary supplementation with n-3 fatty acids from weaning limits brain biochemistry and behavioural changes elicited by prenatal exposure to maternal inflammation in the mouse model. Transl Psychiatry 5: e641. doi: 10.1038/tp.2015.126
[78]	Labrousse VF, Nadjar A, Joffre C, et al. (2012) Short-term long chain Omega3 diet protects from neuroinflammatory processes and memory impairment in aged mice. PLoS One 7: e36861. doi: 10.1371/journal.pone.0036861
[79]	Kerr M (2008) Neurodevelopmental delays associated with iron-fortified formula for healthy infants. Med Psychiatry Mental Health.
[80]	Stein Z, Susser M, Saenger G, et al. (1975) Famine and human development: The Dutch hunger winter of 1944–1945. Q Rev Biol 7: 1944–1945.
[81]	Hoek HW, Susser E, Buck KA, et al. (1996) Schizoid personality disorder after prenatal exposure to famine. Am J Psychiatry 153: 1637–1639. doi: 10.1176/ajp.153.12.1637
[82]	Susser E, Neugebauer R, Hoek HW, et al. (1996) Schizophrenia after prenatal famine: Further evidence. Arch Gen Psychiatry 53: 25–31. doi: 10.1001/archpsyc.1996.01830010027005
[83]	Susser E, Hoek HW, Brown A (1998) Neurodevelopmental Disorders after Prenatal Famine: The Story of the Dutch Famine Study. Am J Epidemiol 147: 214–216.
[84]	Millichap JG, Yee MM (2012) The Diet Factor in Attention-Deficit/hyperactivity disorder. Paediatrics 129: 330. doi: 10.1542/peds.2011-2199
[85]	Surén P, Roth C, Bresnahan M, et al. (2013) Association between maternal use of folic acid supplements and risk of autism spectrum disorders in children. J Am Med Assoc 309: 570–577. doi: 10.1001/jama.2012.155925
[86]	Grant WB, Soles CM (2009) Epidemiologic evidence supporting the role of maternal vitamin D deficiency as a risk factor for the development of infantile autism. Dermatoendocrinology 1: 223–228. doi: 10.4161/derm.1.4.9500
[87]	Blaylock RL (2009) Possible central mechanism in autism spectrum disorders, part 3: The role of excitotoxin food additives and the synergistic effects of other environmental toxins. Altern Ther Health Med 15: 56–60.
[88]	Macfabe DF, Cain DP, Rodriguez-Capote K, et al. (2007) Neurobiological effects of intraventricular propionic acid in rats: Possible role of short chain fatty acids on the pathogenesis and characteristics of autism spectrum disorders.Behav Brain Res176: 149–169.
[89]	Macfabe DF, Cain NE, Boon F, et al. (2011) Effects of the enteric bacterial metabolic product propionic acid on object-directed behavior, social behavior, cognition, and neuroinflammation in adolescent rats: Relevance to autism spectrum disorder. Behav Brain Res 217: 47–54. doi: 10.1016/j.bbr.2010.10.005
[90]	El-Ansary AK, Bacha AB, Kotb M (2012) Etiology of autistic features: The persisting neurotoxic effects of propionic acid. J Neuroinflammation 9: 74.
[91]	Bourgeron T (2007) The possible interplay of synaptic and clock genes in autism spectrum disorders. Cold Spring Harb Symp Quant Biol 72: 645–654. doi: 10.1101/sqb.2007.72.020
[92]	Bourgeron T (2009) A synaptic trek to autism. Curr Opin Neurobiol 19: 231–234. doi: 10.1016/j.conb.2009.06.003
[93]	Bourgeron T (2015) From the genetic architecture to synaptic plasticity in autism spectrum disorder. Nat Rev Neurosci 16: 551–563. doi: 10.1038/nrn3992
[94]	Oberman LM, Ifertmiller F, Najib U, et al. (2016) Abnormal echanisms of plasticity and metaplasticity in autism spectrum disorders and fragile X syndrome. J Child Adolesc Psychopharmacol 26: 617–624. doi: 10.1089/cap.2015.0166
[95]	Georgieff MK, Brunette KE, Tran PV (2015) Early life nutrition and neural plasticity. Dev Psychopathol 27: 411–423. doi: 10.1017/S0954579415000061
[96]	Madore C, Nadjar A, Delpech JC, et al. (2014) Nutritional n-3 PUFAs deficiency during perinatal periods alters brain innate immune system and neuronal plasticity-associated genes. Brain Behav Immun 41: 22–31. doi: 10.1016/j.bbi.2014.03.021
[97]	Lafourcade M, Larrieu T, Mato S, et al. (2011) Nutritional omega-3 deficiency abolishes endocannabinoid-mediated neuronal functions. Nat Neurosci 14: 345–350. doi: 10.1038/nn.2736
[98]	Thomazeau A, Boschbouju C, Manzoni O, et al. (2016) Nutritional n-3 PUFA deficiency abolishes endocannabinoid gating of hippocampal long-term potentiation. Cereb Cortex 27: 2571–2579.
[99]	Fatemi SH, Aldinger KA, Ashwood P, et al. (2012) Consensus paper: Pathological role of the cerebellum in autism. Cerebellum 11: 777–807. doi: 10.1007/s12311-012-0355-9
[100]	Harada M, Taki MM, Nose A, et al. (2011) Non-invasive evaluation of the gabaergic/glutamatergic system in autistic patients observed by mega-editing proton MR spectroscopy using a clinical 3 tesla instrument. J Autism Dev Disord 41: 447–454. doi: 10.1007/s10803-010-1065-0
[101]	Blatt GJ, Fatemi SH (2011) Alterations in gabaergic biomarkers in the autism brain: Research findings and clinical implications. Anat Rec 294: 1646–1652. doi: 10.1002/ar.21252
[102]	Hogart A, Leung KN, Wang NJ, et al. (2009) Chromosome 15q11-13 duplication syndrome brain reveals epigenetic alterations in gene expression not predicted from copy number. J Med Genet 46: 86–93.
[103]	Xu LM, Li JR, Huang Y, et al. (2012) Autismkb: An evidence-based knowledgebase of autism genetics. Nucleic Acids Res 40: 1016–1022. doi: 10.1093/nar/gkr1145
[104]	Jamain S, Betancur C, Quach H, et al. (2002) Linkage and association of the glutamate receptor 6 gene with autism. Mol Psychiatry 7: 302–310. doi: 10.1038/sj.mp.4000979
[105]	Yang Y, Pan C (2013) Role of metabotropic glutamate receptor 7 in autism spectrum disorders: A pilot study. Life Sci 92: 149–153. doi: 10.1016/j.lfs.2012.11.010
[106]	Yip J, Soghomonian JJ, Blatt GJ (2007) Decreased GAD67 mrna levels in cerebellar purkinje cells in autism: Pathophysiological implications. Acta Neuropathol 113: 559–568. doi: 10.1007/s00401-006-0176-3
[107]	Aldred S, Moore KM, Fitzgerald M, et al. (2003) Plasma amino acid levels in children with autism and their families. J Autism Dev Disord 33: 93–97. doi: 10.1023/A:1022238706604
[108]	Shinohe A, Hashimoto K, Nakamura K, et al. (2006) Increased serum levels of glutamate in adult patients with autism. Prog Neuropsychopharmacology Biol Psychiatry 30: 1472–1477. doi: 10.1016/j.pnpbp.2006.06.013
[109]	Sutcliffe JS, Delahanty RJ, Prasad HC, et al. (2005) Allelic heterogeneity at the serotonin transporter locus (slc6a4) confers susceptibility to autism and rigid-compulsive behaviors. Am J Hum Genet 77: 265–279.
[110]	Levitt P (2011) Serotonin and the autisms: A red flag or a red herring? Arch Gen Psychiatry 68: 1093–1094. doi: 10.1001/archgenpsychiatry.2011.98
[111]	Muller CL, Anacker AM, Veenstravanderweele J (2016) The serotonin system in autism spectrum disorder: From biomarker to animal models. Neuroscience 321: 24–41. doi: 10.1016/j.neuroscience.2015.11.010
[112]	Mccracken JT, Mcgough J, Shah B, et al. (2002) Risperidone in children with autism and serious behavioral problems. N Engl J Med 347: 314–321. doi: 10.1056/NEJMoa013171
[113]	Pavăl D (2017) A Dopamine Hypothesis of Autism Spectrum Disorder. Dev Neurosci 39: 355–360. doi: 10.1159/000478725
[114]	Nakamura K, Sekine Y, Ouchi Y, et al. (2010) Brain serotonin and dopamine transporter bindings in adults with high-functioning autism. Arch Gen Psychiatry 67: 59–68. doi: 10.1001/archgenpsychiatry.2009.137
[115]	Kałuzna-Czaplińska J, Socha E, Rynkowski J (2010) Determination of homovanillic acid and vanillylmandelic acid in urine of autistic children by gas chromatography/mass spectrometry. Med Sci Monit 16: CR445–CR450.
[116]	Anderson GH, Johnston JL (1983) Nutrient control of brain neurotransmitter synthesis and function. Can J Physiol Pharmacol 61: 271–281. doi: 10.1139/y83-042
[117]	Fernstrom JD, Fernstrom MH (2007) Tyrosine, phenylalanine, and catecholamine synthesis and function in the brain. J Nutr 137: 1539S–1547S. doi: 10.1093/jn/137.6.1539S
[118]	Schmidt RJ, Tancredi DJ, Krakowiak P, et al. (2014) Maternal intake of supplemental iron and risk of autism spectrum disorder. Am J Epidemiol 180: 890–900. doi: 10.1093/aje/kwu208
[119]	Eyles DW, Burne TH, Mcgrath JJ (2013) Vitamin D, effects on brain development, adult brain function and the links between low levels of vitamin D and neuropsychiatric disease. Front Neuroendocrinol 34: 47–64. doi: 10.1016/j.yfrne.2012.07.001
[120]	Whitehouse AJ, Holt BJ, Serralha M, et al. (2013) Maternal vitamin D levels and the autism phenotype among offspring. J Autism Dev Disord 43: 1495. doi: 10.1007/s10803-012-1676-8
[121]	Tylavsky FA, Kocak M, Murphy LE, et al. (2015) Gestational vitamin 25(OH)D status as a risk factor for receptive language development: A 24-month, longitudinal, observational study. Nutrients 7: 9918–9930. doi: 10.3390/nu7125499
[122]	Morales E, Guxens M, Llop S, et al. (2012) Circulating 25-hydroxyvitamin D3 in pregnancy and infant neuropsychological development. Pediatrics 130: e913–e920. doi: 10.1542/peds.2011-3289
[123]	Keim SA, Bodnar LM, Klebanoff MA (2014) Maternal and cord blood 25(OH)-vitamin D concentrations in relation to child development and behaviour. Paediatr Perinat Epidemiol 28: 434–444. doi: 10.1111/ppe.12135
[124]	Magnusson C, Rai D, Goodman A, et al. (2012) Migration and autism spectrum disorder: Population-based study. Br J Psychiatry 201: 109–115. doi: 10.1192/bjp.bp.111.095125
[125]	Holick MF (2007) Vitamin D deficiency. N Engl J Med 357: 266–281. doi: 10.1056/NEJMra070553
[126]	Vinkhuyzen AA, Eyles DW, Burne TH, et al. (2016) Gestational vitamin D deficiency and autism-related traits: The Generation R Study. Mol Psychiatry 23: 240–246.
[127]	Liu X, Liu J, Xiong X, et al. (2016) Correlation between Nutrition and Symptoms: Nutritional Survey of Children with Autism Spectrum Disorder in Chongqing, China. Nutrients 8: 294. doi: 10.3390/nu8050294
[128]	Egan AM, Dreyer ML, Odar CC, et al. (2013) Obesity in young children with autism spectrum disorders: Prevalence and associated factors. Child Obes 9: 125–131. doi: 10.1089/chi.2012.0028
[129]	Adams JB, Vogelaar AT, (2005) Nutritional abnormalities in autism and effects of nutritional supplementation, In: ASA's 36th National Conference on Autism Spectrum Disorders, Nashville, TN.
[130]	Strambi M, Longini M, Hayek J, et al. (2006) Magnesium profile in autism. Biol Trace Elem Res 109: 97–104. doi: 10.1385/BTER:109:2:097
[131]	Pineless SL, Avery RA, Liu GT (2010) Vitamin B12 optic neuropathy in autism. Pediatrics 126: e967–e970. doi: 10.1542/peds.2009-2975
[132]	Gong ZL, Luo CM, Wang L, et al. (2014) Serum 25-hydroxyvitamin D levels in Chinese children with autism spectrum disorders. Neuroreport 25: 23–27.
[133]	Kocovska E, Andorsdottir G, Weihe P, et al. (2014) Vitamin d in the general population of young adults with autism in the faroe islands. J Autism Dev Disord 44: 2996–3005. doi: 10.1007/s10803-014-2155-1
[134]	Filipek PA, Juranek J, Nguyen MT, et al. (2004) Relative carnitine deficiency in autism. J Autism Dev Disord 34: 615–623. doi: 10.1007/s10803-004-5283-1
[135]	Adams JB (2013) Summary of Dietary, Nutritional, and Medical Treatments for Autism-based on over 150 published research studies. Autism Reasearch Institute Publication 40-Version. Available from: http://autism.asu.edu.
[136]	Berry RC, Novak P, Withrow N, et al. (2015) Nutrition Management of Gastrointestinal Symptoms in Children with Autism Spectrum Disorder: Guideline from an Expert Panel. J Acad Nutr Diet 115: 1919–1927. doi: 10.1016/j.jand.2015.05.016
[137]	Stewart PA, Hyman SL, Schmidt BL, et al. (2015) Dietary Supplementation in Children with Autism Spectrum Disorders: Common, Insufficient, and Excessive. J Acad Nutr Diet 115: 1237–1248. doi: 10.1016/j.jand.2015.03.026
[138]	Santosh PJ, Singh J (2016) Drug treatment of autism spectrum disorder and its comorbidities in children and adolescents. BJ Psych Adv 22: 151–161.
[139]	Brondino N, Fusar-Poli L, Panisi C, et al. (2016) Pharmacological Modulation of GABA Function in Autism Spectrum Disorders: A Systematic Review of Human Studies. J Autism Dev Disord 46: 825–839. doi: 10.1007/s10803-015-2619-y
[140]	Kumar B, Prakash A, Sewal RK, et al. (2012) Drug therapy in autism: A present and future perspective. Pharmacol Rep 64: 1291–1304. doi: 10.1016/S1734-1140(12)70927-1
[141]	Doyle CA, Mcdougle CJ (2012) Pharmacologic treatments for the behavioral symptoms associated with autism spectrum disorders across the lifespan. Dialogues Clin Neurosci 14: 263–279.
[142]	Croen LA, Grether JK, Yoshida CK, et al. (2011) Antidepressant use during pregnancy and childhood autism spectrum disorders. Arch Gen Psychiatry 68: 1104–1112. doi: 10.1001/archgenpsychiatry.2011.73
[143]	Aman MG, Arnold LE, Mcdougle CJ, et al. (2005) Acute and long-term safety and tolerability of risperidone in children with autism. J Child Adolesc Psychopharmacol 15: 869–884. doi: 10.1089/cap.2005.15.869
[144]	Mcdougle CJ, Scahill L, Aman MG, et al. (2005) Risperidone for the core symptom domains of autism: Results from the study by the autism network of the research units on pediatric psychopharmacology. Am J Psychiatry 16: 1142–1148.
[145]	Buitelaar JK, Willemsen-Swinkels SHN (2000) Medication treatment in subjects with autistic spectrum disorders. Eur Child Adolesc Psychiatry 9: S85–S97. doi: 10.1007/s007870070022
[146]	Aoki Y, Yahata N, Watanabe T, et al. (2014) Oxytocin improves behavioural and neural deficits in inferring others' social emotions in autism. Brain 137: 3073–3086. doi: 10.1093/brain/awu231
[147]	Watanabe T, Abe O, Kuwabara H, et al. (2014) Mitigation of sociocommunicational deficits of autism through oxytocin-induced recovery of medial prefrontal activity: A randomized trial. JAMA Psychiatry 71: 166–175. doi: 10.1001/jamapsychiatry.2013.3181
[148]	Geraghty ME, Bates-Wall J, Ratliff-Schaub K, et al. (2010) Nutritional interventions and therapies in autism: A spectrum of what we know: Part 2. Ican Infant Child Adolesc Nutr 2: 120–133.
[149]	Geraghty ME, Depasquale GM, Lane AE (2010) Nutritional intake and therapies in autism: A spectrum of what we know: Part 1. Ican Infant Child Adolesc Nutr 2: 62–69. doi: 10.1177/1941406409358437
[150]	Marti LF (2014) Dietary interventions in children with autism spectrum disorders-an updated review of the research evidence. Curr Clin Pharmacol 9: 335–349. doi: 10.2174/15748847113086660074
[151]	Kawicka A, Regulska-Ilow B (2013) How nutrition status, diet and dietary supplements can affect autism. A review. Rocz Panstw Zakl Hig 64: 1–12.
[152]	Curtis LT, Patel K (2008) Nutritional and environmental approaches to preventing and treating autism and attention deficit hyperactivity disorder (ADHD): A review. J Altern Complement Med 14: 79–85. doi: 10.1089/acm.2007.0610
[153]	Kidd PM (2002) Autism, an extreme challenge to integrative medicine. Part II: Medical Management. Altern Med Rev 7: 472–499.
[154]	Kidd PM (2002) Autism, an extreme challenge to integrative medicine. Part 1: The knowledge base. Altern Med Rev 7: 292–316.
[155]	Pilla SSDD, Ravisankar P, Penugonda V, et al. (2014) Dietary interventions in Autism Spectrum Disorders. AP J Psychol Med 15: 24–31.
[156]	Joanna KCE, Rynkowski JB (2011) Vitamin supplementation reduces excretion of urinary dicarboxylic acids in autistic children. Nutr Res 31: 497–502. doi: 10.1016/j.nutres.2011.06.002
[157]	Mousainbosc M, Roche M, Polge A, et al. (2006) Improvement of neurobehavioral disorders in children supplemented with magnesium-vitamin B6. II. Pervasive developmental disorder-autism. Magnesium Res 19: 53–62.
[158]	Adams JB, Holloway C (2004) Pilot study of a moderate dose multivitamin/mineral supplement for children with autistic spectrum disorder. J Altern Complement Med 10: 1033–1039. doi: 10.1089/acm.2004.10.1033
[159]	Adams JB, Audhya T, Mcdonoughmeans S, et al. (2011) Effect of a vitamin/mineral supplement on children and adults with autism. BMC Pediatr 11: 111. doi: 10.1186/1471-2431-11-111
[160]	Cannell JJ (2008) Autism and vitamin D. Med Hypotheses 70: 750–759. doi: 10.1016/j.mehy.2007.08.016
[161]	Jia F, Wang B, Shan L, et al. (2015) Core symptoms of autism improved after vitamin D supplementation. Pediatrics 135: e196–e198. doi: 10.1542/peds.2014-2121
[162]	Saad K, Abdel-Rahman AA, Elserogy YM, et al. (2018) Randomized controlled trial of vitamin D supplementation in children with autism spectrum disorder. J Child Psychol Psychiatry 59: 20–29. doi: 10.1111/jcpp.12652
[163]	Jia F, Shan L, Wang B, et al. (2017) Bench to bedside review: Possible role of vitamin D in autism spectrum disorder. Psychiatry Res 260: 360–365.
[164]	Politi P, Cena H, Emanuele E (2011) Dietary Supplementation of Omega-3 Polyunsaturated Fatty Acids in Autism. Handb Behav Food Nutr 88: 1787–1796.
[165]	Malow BA, Adkins KW, Mcgrew SG, et al. (2012) Melatonin for sleep in children with autism: A controlled trial examining dose, tolerability, and outcomes. J Autism Dev Disord 42: 1729–1737. doi: 10.1007/s10803-011-1418-3
[166]	Reading R (2011) Melatonin in autism spectrum disorders: A systematic review and meta-analysis. Dev Med Child Neurol 53: 783–792. doi: 10.1111/j.1469-8749.2011.03980.x
[167]	Critchfield JW, Van HS, Ash M, et al. (2011) The potential role of probiotics in the management of childhood autism spectrum disorders. Gastroenterol Res Pract 2011: 161358.
[168]	Hsiao EY,Mcbride SW, Hsien S, et al. (2013) The microbiota modulates gut physiology and behavioral abnormalities associated with autism. Cell 155: 1451–1463. doi: 10.1016/j.cell.2013.11.024
[169]	Shaaban SY, El Gendy YG, Mehanna NS, et al. (2017) The role of probiotics in children with autism spectrum disorder: A prospective, open-label study. Nutr Neurosci 2017: 1–6.
[170]	Gvozdjáková A,Kucharská J, Ostatníková D, et al. (2014) Ubiquinol improves symptoms in children with autism. Oxid Med Cell Longevity 2014: 798957.
[171]	Geier DA, Kern JK, Davis G, et al. (2011) A prospective double-blind, randomized clinical trial of levocarnitine to treat autism spectrum disorders. Med Sci Monit 17: P115–P123.
[172]	Hardan AY, Fung LK, Libove RA, et al. (2012) A randomized controlled pilot trial of oral N-acetylcysteine in children with autism. Biol Psychiatry 71: 956–961. doi: 10.1016/j.biopsych.2012.01.014
[173]	Chez MG, Buchanan CP, Aimonovitch MC, et al. (2002) Double-blind, placebo-controlled study of L-carnosine supplementation in children with autistic spectrum disorders. J Child Neurol 17: 833–837. doi: 10.1177/08830738020170111501
[174]	Elder JH (2008) The gluten-free, casein-free diet in autism: An overview with clinical implications. Nutr Clin Pract 23: 583–588. doi: 10.1177/0884533608326061
[175]	Mulloy TA, Lang R, Reilly OM, et al. (2009) Gluten-free and casein-free diets in the treatment of autism spectrum disorders: A systematic review. Res Autism Spectrum Disord 4: 328–339.
[176]	Cade R, Privette M, Fregly M, et al. (2000) Autism and Schizophrenia: Intestinal disorders. Nutr Neurosci 3: 57–72. doi: 10.1080/1028415X.2000.11747303
[177]	Whiteley P, Haracopos D, Knivsberg AN, et al. (2010) The ScanBrit randomized, controlled, single-blind study of a gluten- and casein- free dietary intervention for children with autism spectrum disorders. Nutr Neurosci 13: 87–100. doi: 10.1179/147683010X12611460763922
[178]	Evangeliou A, Vlachonikolis I, Mihailidou H, et al. (2003) Application of a ketogenic diet in children with autistic behavior: Pilot study. J Child Neurol 18: 113–118. doi: 10.1177/08830738030180020501
[179]	Rossignol DA (2009) Novel and emerging treatments for autism spectrum disorders: A systematic review. Ann Clin Psychiatry 21: 213–236.
[180]	Elrashidy O, Elbaz F, Elgendy Y, et al. (2017) Ketogenic diet versus gluten free casein free diet in autistic children: A case-control study. Metab Brain Dis 32: 1935–1941. doi: 10.1007/s11011-017-0088-z
[181]	Strickland E (2009) Eating for Autism: The 10-step Nutrition Plan to Treat Your Child's Autism, Asperger's, or ADHD. Philadelphia, PA: Da Capo Press.

This article has been cited by:

1.	Carlota Quiban, Raelene Brooks, Dometrives Armstrong, Caring for Adult Patients With Autism in the Critical Care Setting, 2020, 43, 0887-9303, 58, 10.1097/CNQ.0000000000000291
2.	Anna Błażewicz, Iwona Szymańska, Aleksander Astel, Agnieszka Stenzel-Bembenek, Wojciech Remington Dolliver, Agata Makarewicz, Assessment of Changes over Time of Lipid Profile, C-Reactive Protein Level and Body Mass Index in Teenagers and Young Adults on Different Diets Belonging to Autism Spectrum Disorder, 2020, 12, 2072-6643, 2594, 10.3390/nu12092594
3.	Arefeh Kardani, Amin Soltani, Robert D.E. Sewell, Mehrdad Shahrani, Mahmoud Rafieian-Kopaei, Neurotransmitter, Antioxidant and Anti-neuroinflammatory Mechanistic Potentials of Herbal Medicines in Ameliorating Autism Spectrum Disorder, 2020, 25, 13816128, 4421, 10.2174/1381612825666191112143940
4.	Sharmistha Saha, Tanusree Saha, Usha Rajamma, Swagata Sinha, Kanchan Mukhopadhyay, Analysis of association between components of the folate metabolic pathway and autism spectrum disorder in eastern Indian subjects, 2022, 49, 0301-4851, 1281, 10.1007/s11033-021-06956-z
5.	Olakunle James Onaolapo, Adejoke Yetunde Onaolapo, Nutrition, nutritional deficiencies, and schizophrenia: An association worthy of constant reassessment, 2021, 9, 2307-8960, 8295, 10.12998/wjcc.v9.i28.8295
6.	Xiao Li, The Relationship Between Dietary Patterns and ASD, 2022, 19, 2791-0210, 73, 10.54097/hset.v19i.2697
7.	Sara Almsmary, Waala Alwarfaly, Selima Muftah, Ashmisa Ebrahim Eltuhami, Zahzahan Ateia Alsaeti, Reima Mansour, Ali Ateia Elmabsout, Childhood Autism: Clinical Characteristics, Nutritional Status, and Psychosocial Features, 2022, 4, 2593-8339, 98, 10.24018/ejmed.2022.4.3.1156
8.	Olakunle J. Onaolapo, Ademola O. Odeniyi, Adejoke Y. Onaolapo, Parkinson’s Disease: Is there a Role for Dietary and Herbal Supplements?, 2021, 20, 18715273, 343, 10.2174/1871527320666210218082954
9.	Sharmistha Saha, Mahasweta Chatterjee, Nilanjana Dutta, Swagata Sinha, Kanchan Mukhopadhyay, GABA Receptor SNPs and Elevated Plasma GABA Levels Affect the Severity of the Indian ASD Probands, 2022, 72, 0895-8696, 1300, 10.1007/s12031-022-02023-9

Reader Comments

Your name:*

Email:*
© 2018 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

AIMS Medical Science

0.7

Metrics

Article views(12658) PDF downloads(2160) Cited by(9)

Preview PDF

Download XML

Export Citation

Article outline

Show full outline

AIMS Medical Science

Nutrition in autism spectrum disorders: A review of evidences for an emerging central role in aetiology, expression, and management

Related Papers:

Abstract

1. Introduction

2. Conceptual background and hypotheses

2.1. Trading volume and its relation to unscheduled events

2.2. Bitcoin addresses and information asymmetry

3. Data and methods

3.1. Data and sample

3.2. Dependent variables and event study methodology

3.3. Independent and control variables

4. Results

4.1. Descriptive statistics

4.2. Empirical results

4.2.1. Abnormal trading volume around large Bitcoin transactions

4.2.2. Abnormal trading volume and effects of information asymmetry

6. Concluding remarks

Acknowledgments

Conflicts of interest

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Other Articles By Authors

Catalog

AIMS Medical Science

Nutrition in autism spectrum disorders: A review of evidences for an emerging central role in aetiology, expression, and management

Related Papers:

Abstract

1. Introduction

2. Conceptual background and hypotheses

2.1. Trading volume and its relation to unscheduled events

2.2. Bitcoin addresses and information asymmetry

3. Data and methods

3.1. Data and sample

3.2. Dependent variables and event study methodology

3.3. Independent and control variables

4. Results

4.1. Descriptive statistics

4.2. Empirical results

4.2.1. Abnormal trading volume around large Bitcoin transactions

4.2.2. Abnormal trading volume and effects of information asymmetry

6. Concluding remarks

Acknowledgments

Conflicts of interest

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Other Articles By Authors

Related pages

Tools

Export File

Citation

Format

Content

Catalog